AP42 Section: 7.1

Background Chapter 5

Reference: 8

Title: Hydrocarbon Emissions Floating Roof Petroleum Tanks

Engineering Science, Inc., prepared for Western Oil and Gas Association,

Los Angeles, CA,

January 1977

HYDROCARBON EMISSIONS FROM FLOATING ROOF PETROLEUM TANKS

ENGINEERING-SCIENCE, INC.

HYDROCARBON E M I S S I O N S FROM FLOATING-ROOF STORAGE TANKS

PREPARED FOR THE WESTERN OIL AN0 GAS A S S O C I A T I O N

609 SOUTH GRANO AVENUE LOS ANGELES, C A L I F O R N I A 91107

JANUARY 1917

SUBMITTED BY

ENGINEER ING-SCI ENCE, I N C . 150 NORTH SANTA ANITA AVENUE A R C A O I A . CALIFORNIA B l O O B

ENQINEERINQ-SCIENCE - FOREWORD

The r e l a t i o n s h i p of hydrocarbon emissions t o the photochemical

ox idant p o l l u t i o n problcms of many urban a reas of t h e United S t a t e s

is not w e l l understood. S t i l l , many r egu la to ry agencies have taken

t h e pos i t i on t h a t hydrocarbon emissions from a l l sources should be

reduced in order t o minimize t h e ex ten t of t h e oxidant problem.

With regard t o hydrocarbon emissions from f loat ing-roof s to rage

tanks, the only b a s i s a v a i l a b l e t o t h e r egu la to ry agencies f o r esti-

mating such emissions a t t h e p re sen t t i m e is a document published by

t h e American Petroleum I n s t i t u t e i n February, 1962 e n t i t l e d , - "API

B u l l e t i n on Evaporation Loss From Floating-Roof Tanks" and i d e n t i f i e

a s MI-2517. The Western O i l and Gas Associat ion undertook t h i s

i n v e s t i g a t i o n t o measure a c t u a l hydrocarbon los ses from s e v e r a l tanks

and t o determine whether or not t h e API equat ion properly est imated

hydrocarbon emissions from f loa t ing- roof s to rage tanks; t hese o b j e c t i v e s

were m e t . It i s hoped t h a t t h i s r e p o r t w i l l s e rve a s a u s e f u l guide

t o t h e r egu la to ry agencies - and t o o t h e r s - i n determining t h e need

f o r and t h e d i r e c t i o n of f u t u r e r e sea rch concerning hydrocarbon

emissions from f loa t ing- roof s to rage tanks.

i

ENOINEERINO-SCIENCE - ACKNOWLEDGMENTS

The Western Oil and Gas Assoc ia t ion (WOGA) cont rac ted w i t h

Engineering-Science, Inc. (ES), f o r t h i s eva lua t ion of hydrocarbon

emissions from f loat ing-roof s t o r a g e tanks in J u l y , 1976. Chairmen

of WOGA committees who conceived and approved t h i s p r o j e c t were:

D r . Car le ton B. S c o t t , Chairman, WOGA Environmental Conservation Committee (ECC), Union Oil Company of C a l i f o r n i a .

M r . W i l l i a m B. Harrell, Chairman, WOGA P r o j e c t Coordinating Subcommittee, Mobil Oil Corporat ion.

While t h e ECC has o v e r a l l r e s p o n s i b i l i t y f o r WOGA's environmental

programs, s p e c i f i c p r o j e c t s such a s t h i s one a r e c a r r i e d out under

the d i r e c t i o n of appointed t a sk groups of engineers and s c i e n t i s t s .

each possessing q u a l i f i c a t i o n s in some phase of t h e s u b j e c t under

cons idera t ion . WCGA d i r e c t e d and monitored t h i s p r o j e c t through the

fol lowing s p e c i a l l y appointed t a s k force :

Name

P e t e r E. Jonker , Chairman W i l l i a m J. P o r t e r D r . Robert L. Russe l l P e t e r L. Mehta Earl K. Dewey, Jr. . J e r r y Adams John A. Glaser Richard A. O ' H a r e Gordon J. Good Robert M. Stoneham Hayden H. Jones

Organizat ion

Union Oil Company of C a l i f o r n i a Chevron U.S.A. Incorporated Union Oil Research A t l a n t i c Richf ie ld Company Cont inenta l Oil Company F l e t c h e r O i l Ref ining Company Gulf Oil Company, U.S. S h e l l Oil Company Standard O i l Company of Ohio Texaco, Incorporated Union Oil Company of C a l i f o r n i a

ii

ENQINEERINQ-SCIENCE -

The success fu l completion of t h i s p r o j e c t was p a r t i a l l y ensured

by t h e unbiased e f f o r t s and open d ia logue of s e v e r a l engineers and

s c i e n t i s t s from t h e a i r p o l l u t i o n con t ro l agencies . Open meetings

were held monthly t o d i s c u s s progress and a l l i d e a s voiced by these

p a r t i c i p a n t s were f u l l y considered within t h e limits of e s t ab l i shed

t i m e and budget c o n s t r a i n t s . The following o rgan iza t ions were re- presented a t d i f f e r e n t t imes and by d i f f e r e n t i nd iv idua l s throughout

t h e course of t h i s i n v e s t i g a t i o n :

Environmental P r o t e c t i o n Agency Research Tr i ang le Park, North Carol ina

Environmental P r o t e c t i o n Agency Region I X , San Franc isco

C a l i f o r n i a A i r Resources Board Sacramento, C a l i f o r n i a

Southern C a l i f o r n i a A i r P o l l u t i o n Control D i s t r i c t Los Angeles, C a l i f o r n i a

San Diego A i r P o l l u t i o n Control D i s t r i c t San Diego, C a l i f o r n i a

Suggestions and ideas a l s o were provided by:

American Petroleum I n s t i t u t e Committee on Evaporat ive L o s s Measurements Washington, D.C.

Supporting Engineering-Science, Inc. in s p e c i a l t echn ica l a r e a s

were:

DK. W. L. F a i t h , Consulting Engineer San Marino, C a l i f o r n i a

Ana ly t i ca l Research Labora tor ies Monrovia, C a l i f o r n i a

The P r o j e c t Manager f o r t h i s i n v e s t i g a t i o n w a s M r . A. L. Wilson,

and t h e Technical DirectOK w a s M r . M. Dean High, Vice P res iden t ,

Engineering-Science, Inc. To a l l who cont r ibu ted t o t h i s s tudy and

p repa ra t ion of t h i s r e p o r t , Engineering-Science, Inc . extends i t s

s i n c e r e apprec i a t ion .

iii

ENQINEERINQ-SCIENCE -

FOREWORD

ACKNOWLEDGMENT

CHAPTER I

CHAPTER I1

CHAPTER I11

CHAPTER I V

CHAPTER V

CHAPTER V I

CHAPTER V I 1

CHAPTER V I 1 1

CHAPTER I X

APPENDICES

TABLE OF CONTENTS

EXECUTIVE S W R Y

CONCLUSIONS

INTRODUCTION

Background Objec t ive Scope Methodology

MEASUREMENTS

Sample Co l l ec t ion Procedure Laboratory Procedure Evaluat ion of P o t e n t i a l S t r a t i f i c a -

t i o n in t h e Tanks E f f e c t s of Water, Temperature and

Flash ing T i m e Exclusion of C e r t a i n Tanks Densi ty Measurements

PARAMETERS

General Comparisons wi th API Equation Cor re l a t ion With Measured Parameters

FURTHER DISCUSSION OF SIGNIFICANT

STATIC STORAGE HYDROCARBON WISSION

CORRELATION OF EMISSIONS WITH MEASURED

PARAMETERS

BEST AVAILABLE SEAL TECHNOLOGY

General Survey of Avai lab le Sea l s

Primary Sea l s Secondary S e a l s S e a l and Tank Accessories

Survey of I n s t a l l e d Sea l s with Respect t o Gap S ize

Discussion of Resu l t s

RECOMMENDATIONS

BIBLIOGRAPHY

UNDER SEPARATE COVER

i

ii

1-1

11-1

111-1

111-1 111-2 111-3 111-5

1v-1

1v-2 1v-3

1v-9

1v-14 1v-15 1v-15

v-1 v-1 v-3 V-6

v1-1

v11-1

v11-1 v11-2 v11-2 v11-8 VII-12

VII-19 V I I - 2 8

V I I I - 1

1x-1

ENQINEERINQ-SCIENCE PI-

No. - 1-1

I V - 1

IV-2 IV-3 IV-4 IV-5 IV-6 IV-7

v- 1

v-2

v-3

v-4

v-5

V-6

V I I - 1 VII-2

V I I - 3 VII-4 VII-5 VII-6 VII -7 VII-8 VII-9 V I I - 1 0 V I I - 1 1 VII-12 VII-13

LIST OF FIGURES

T i t l e

Descr ip t ion of 13 Floating-Roof Tanks Used 1-2 in t h e Study (Excludes Union No. 5 and 3 Crude O i l Tanks)

Example Density-Evaporation Curve, Arco #134, IV-7 September 8 , 1976.

Summary of Tank Test (Union) IV-10 Summary of Tank Test (Texaco) I V - 1 1

IV-16 IV-17

Change i n Stock Densi ty with Time I V - 1 8 Change i n Stock Densi ty wi th Time IV-19

Comparison of API Calcula ted and Measured v-5

Hydrocarbon Emission Rate as a Function of v-7

Hydrocarbon Emission Rate a s a Function of V- 8

Hydrocarbon Emission Rate a s a Funct ion of v-9

Hydrocarbon Emission Rate Compared t o Age of v-10

Change in Stock Densi ty wi th T i m e Change in Stock Densi ty w i t h Time

Emission Rates

Reid Vapor Pressure and Tank Diameter

Average Wind Speed and Average Ambient Temperature

Sea l Gap Arga and Number of Gaps

Primary Sea l

Temperature t o Change i n Densi ty Standard 192-Comparison of Wind Speed and Stock V-13

Weight Ass is ted M e t a l l i c Shoe Sea l SR-7 Horton Fabr i c S e a l ( R e s i l i e n t Type)

PDM Tubeseal System Kisses1 - Thetachron Tank Secondary Wiper S e a l U l t r a f l o t e Wiper P a c i f i c E r e c t o r s Mini Bag Atlas Mini Bag Tr i co Mini Bag Tr i co Wiper Maloney Secondary Tank S e a l Spring A s s i s t e d Mechanism

f o r Horton F l o a t i n g Roofs

VII-3 VII-5

VII-6 V I I - 7 VII-9 VII-10 V I I - 1 1 VII-13 V I I - 1 4 VII-15 VII-16 V I I - 1 7 VII-18

ENQINEERING-SCIENCE -

No.

111-1

-

I V - 1

IV-2 IV-3

IV-4 IV-5 IV-6 IV-7 I V - 8 IV-9 IV-10 I V - 1 1 IV-12 IV-13 IV-14 IV-15 IV-16 IV-17 IV-18 IV-19 IV-20 IV-21

v-1 v- 2

v-3

V I - 1

V I I - 1 VII-2

LIST OF TABLES

T i t l e - Descr ip t ion of Floating-Roof Tanks Used f o r

Study

Ten ta t ive Method f o r Determining Evaporation

Densi ty and Spec i f i c Gravi ty of Liquids by

Example Depth S t r a t i f i c a t i o n Resul t s A Comparison of Densi ty Measurements Wide Mouth B o t t l e s Versus Small Mouth B o t t l e s

F i e l d Measurement Summary (Arco) F i e l d Measurement Summary (Exxon) F i e l d Measurement Sununary (Gulf) F i e ld Measurement Summary (Lion) F ie ld Measurement Sumnary (Mobil) F i e l d Measurement Summary (She l l ) F i e l d Measurement Summary (She l l ) F i e l d Measurement Summary (Standard) F i e l d Measurement Summary (Standard) F i e l d Measurement Summary (Standard) F i e l d Measurement Summary (Standard) F i e l d Measurement Summary (Standard F i e l d Measurement Summary (Texaco) F i e l d Measurement Sumnary (Texaco) F i e l d Measurement Summary (Union) F i e l d Measurement Summary (Union) F i e l d Measurement Summary (Union) Measured Emission Rate

Standing Storage Loss Emissions Summary API-2517 Calculated Versus Observed Hydrocarbon

Losses by Change i n Stock Densi ty

Bingham Pycnometer

Emissions

111-4

1v-4

IV-4 . iv-12

1v-13 iv-21 1v-22 1v-23 1v-24 1v-25 1v-26 1v-27 IV-28 : 1v-29 1v-30 1v-31 1v-32 1v-33 1v-34 1v-35 1v-36 1v-37 1v-38

v-2

v-4 Comparison of Measured Gaps Versus Allowable Gaps on T e s t Tanks V-14

Comparison. of Engineering-Science F i e l d Test Var iab les t o AFT-2517 Equation Var iab les v1-2

Performance of S e a l s on Floating-Roof Tanks (W) VII-20 Performance of Sea l s on Floating-Roof T h k s (R) V I I - 2 4

CHAPTER I

E X E C U T I V E SUMMARY

. . .

ENGINEERING-SCIENCE - CHAPTER I

EXECUTIVE SUMMARY

For the p a s t 15 years o r so, AF'I B u l l e t i n 2517 has been used t o

estimate hydrocarbon l o s s e s from f loa t ing- roof s to rage tanks. In

r ecen t years , much a t t e n t i o n has been focussed on t h e sources of

hydrocarbons and t h e r egu la to ry agencies have pushed a l l sources f o r

a p p l i c a t i o n of t h e b e s t a v a i l a b l e c o n t r o l technology. In t h e case of

f loa t ing- roof tanks, no r e c e n t measurement d a t a e x i s t e d on the hydro-

carbon l o s s e s a s soc ia t ed wi th d i f f e r e n t tanks , s e a l s , s tock con ten t s ,

o r weather parameters.

This i n v e s t i g a t i o n was i n i t i a t e d p r imar i ly t o provide c u r r e n t

hydrocarbon emission d a t a on f u l l - s i z e s t o r a g e tanks in Cal i fo rn ia .

Seventeen tanks were o r i g i n a l l y made a v a i l a b l e but only 13 tanks con-

t a i n i n g d i s t i l l a t e products were eva lua ted f o r the du ra t ion of the

p r o j e c t . A p i c t o r i a l d e s c r i p t i o n of t h e s e 13 tanks and some of t h e

major parameters which a r e commonly a s soc ia t ed with hydrocarbon l o s s e s

from f loa t ing- roof tanks are i l l u s t r a t e d in Figure 1-1. Work was

i n i t i a t e d i n J u l y 1976 and concluded i n January 1977.

To determine hydrocarbon l o s s e s , ES followed a procedure i n API B u l l e t i n 2512 which r e l a t e d hydrocarbon l o s s t o s tock dens i ty

change.

evaporate most qu ick ly from t h e tank w i l l be l i g h t ends and t h e s tock

d e n s i t y w i l l , t h e r e f o r e , i nc rease as t o t a l volume decreases . Samples

of t h e s t o c k were c o l l e c t e d a t p e r i o d i c i n t e r v a l s and d e n s i t y de t e r -

mina t ions were conducted i n t h e l abora to ry . The f l a s h i n g p o t e n t i a l

of t h e petroleum s tocks used i n t h i s s tudy was s u f f i c i e n t l y l a r g e so

as t o overshadow a l l o t h e r poss ib l e e r r o r s a s soc ia t ed wi th both f i e l d

and l abora to ry procedures . , Seve ra l modi f ica t ions t o the gene ra l API

procedure had t o be accomplished.

t r o l l e d , t h e i n v e s t i g a t i o n progressed smoothly.

The method relies on t h e f a c t that hydrocarbon compounds t h a t

Once t h e f l a s h i n g problem was con-

Auxi l ia ry t a s k s a s soc ia t ed w i t h t h i s p r o j e c t included labora tory

and p i l o t s c a l e t e s t i n g , surveys of seal technology, and review of t h e

I- 1

x x x

x x x x x x x x x x

W oz 3 M M W a o

0 0 0 Y ) O O O O NY.-N'-J

x x x x

x x x x x x x x

0 0 0 0 0 c c c c c 0 - 0 0 0

N - 0 0 - N

L 3

r Y P

L f

x x

x x x

Y X

x x X Y

x x

m m m m m m m m - N n - m w - 0 0 0 0 0 0 0 0 c c c c c c c c

0 - 0 0 0 0 0 0 0 - N ' - J W V ) Y ) C

x x x x I . x x x x x x x x

I- 2

F I G U R E I -

x x x x x x x

X ~~

x x x x x x x x x x x x

x x

I x x x x x x - x x x

X x x X I X I X X I x X x

ENGINEER ING-SC IENCE , INC.

ENQINEERING-SCIENCE - API equat ion f o r e s t ima t ing hydrocarbon los ses .

Resu l t s of t h e f i e l d i n v e s t i g a t i o n showed t h a t average hydrocarbon

emissions from 13 s t a t i c f loa t ing- roof tanks were about 50 percent

of t h e q u a n t i t y t h a t would be es t imated by use of API B u l l e t i n 2517.

Severa l tank, roof and s e a l design parameters a s w e l l a s s e v e r a l

weather parameters were recorded dur ing t h e s tudy and s t a t i s t i c a l

c o r r e l a t i o n s were conducted t o t r y t o determine t h e s i g n i f i c a n t v a r i -

a b l e s which a f f e c t e d emission r a t e s . Unfortunately, some v a r i a b l e s

such a s s tock temperature i n t h e r i m space, r i m space su r face a rea ,

t i g h t n e s s of t h e apron between t h e roof and shoe s e a l s , e t c . , were

not measured dur ing t h i s i n v e s t i g a t i o n . As might be expected, no

c o r r e l a t i o n s were p a r t i c u l a r l y ev iden t ; f u r t h e r r e sea rch w i l l be

requi red t o c l e a r l y understand t h e mechanisms r e spons ib l e f o r hydro-

carbon los ses .

Hydrocarbon emissions from 9 of t h e 13 t e s t e d tanks showed sur-

p r i s i n g l y similar emissions even though major d i f f e r e n c e s were noted

between Reid vapor p re s su re of t h e s tocks , s e a l type , seal gaps,

wind speed and o t h e r parameters.

no more importance than many o t h e r parameters. I n one example, two

very similar t anks conta in ing t h e same type of product had thp same

emission rate, al though one had twice t h e gap s i z e

S e a l gap s i z e was found t o be of

P a r t i c u l a r i n t e r e s t dur ing t h i s i n v e s t i g a t i o n was focussed on

determining t h e b e s t a v a i l a b l e seal technology. Two surveys were

conducted. The f i r s t survey eva lua ted commercially a v a i l a b l e s e a l s

through i n d i v i d u a l companies. The second survey addressed seal gap

s i z e a c t u a l l y measured a t d i f f e r e n t roof l e v e l s f o r some 200 t anks

i n Southern Ca l i fo rn ia .

Many seal des igns were found t o be commercially a v a i l a b l e but

t h e r e were no emission test d a t a a v a i l a b l e t o demonstrate t h e c o s t

e f f e c t i v e n e s s of t h e des igns f o r reducing hydrocarbon emissions.

The t a b u l a t i o n of gap s i z e d a t a f o r 200 t anks showed t h a t tube

s e a l s f r equen t ly have smal le r gap spaces than primary shoe seals and

t h a t s e a l gap s i z e could be reduced by use of secondary seals.

1-3

CHAPTER I 1

CONCLUS IONS

ENGINEERING-SCIENCE -

1.

2.

3 .

4 .

CHAPTER 11

CONCLUSIONS

The bas i c API method of determining hydrocarbon l o s s e s from

petroleum d i s t i l l a t e products by measuring t h e s tock dens i ty

change (API-2512) was demonstrated t o be a very r e l i a b l e pro-

cedure. However, t h e ins t rumenta t ion and methodology developed

during t h e course of t h i s i n v e s t i g a t i o n need t o be incorporated

i n t o any re -wr i te of t h e API procedure. The API procedure w a s

found not s a t i s f a c t o r y f o r measuring s tanding l o s s e s from crude

o i l s to rage tanks.

Measured hydrocarbon l o s s e s from 13 f loa t ing- roof s to rage tanks ,

a l l loca ted i n C a l i f o r n i a and a l l conta in ing petroleum d i s - " t i l l a t e s , were i n gene ra l below t h e q u a n t i t i e s es t imated by

t h e MI-2517 equat ion. The emission lo s s of t h e 13 tanks over

t h e du ra t ion of t h e test per iod averaged about one-half of t h e

q u a n t i t y es t imated by API-2517. All measurement procedure de-

c i s i o n s w e r e decided on t h e conse rva t ive ly high s i d e .

Four f loa t ing- roof tanks wi th primary tube seals and zero o r very

small gap s i z e showed t h e same o rde r of magnitude f o r hydrocarbon

l o s s e s as s i x tanks s t o r i n g s i m i l a r products wi th primary shoe

s e a l s having cons iderably more gap space. Two i d e n t i c a l r i v e t e d

tanks wi th t h e same type of primary shoe s e a l s , t h e same s tock ,

and loca ted side-by-side i n t h e same tank farm, had t h e same

emission rate even though t h e t o t a l gap a r e a on one tank was

i n t e n t i o n a l l y increased t o more than double t h e gap a r e a on the

second tank. It would appear t h a t some o t h e r parameters a r e

over-r iding t h e s i g n i f i c a n c e of gap s i z e wi th r e spec t t o hydro-

carbon los ses .

A survey of a wide v a r i e t y of s e a l s on 200 tanks i n Southern

C a l i f o r n i a showed t h a t no s e a l des ign would e l imina te a l l gaps

on f loa t ing- roof tanks.

11-1

CHAPTER 1 1 1

I NTROOUCT I ON

ENGINEERING-SCIENCE - CHAPTER 111

INTRODUCTION

BACKGROUND

Hydrocarbon emissions from f loa t ing- roof tanks a r e c u r r e n t l y

es t imated by u s e of an equat ion from an American Petroleum I n s t i t u t e

(API) b u l l e t i n . The da ta base f o r t h i s b u l l e t i n cons i s t ed of 60 tests

conducted p r imar i ly between 1933 and 1942. Two tests were r u n on

crude o i l ; 58 were run on gaso l ine . The l o s s measurement was determined

i n 57 of t h e t e s t s by t h e change i n vapor pressure. Three-fourths of

t h e tests were run on r i v e t e d tanks and t h e l a r g e s t tank t e s t e d was 145

f e e t i n diameter.

Resu l t s of t h e s e s t u d i e s were publ ished i n 1962 by t h e American

Petroleum I n s t i t u t e i n t h e i r API B u l l e t i n 2517. The r e p o r t included

a n equat ion f o r e s t ima t ing hydrocarbon l o s s e s from f loat ing-roof tanks.

The U.S. Pub l i c Heal th Serv ice (predecessor t o EPA) has used t h e

equat ion and published it i n t h e i r emissions f a c t o r pub l i ca t ion , AP-42.

A i r p o l l u t i o n c o n t r o l agencies a t a l l l e v e l s of government have s i n c e

used t h e equat ion t o e s t ima te hydrocarbon emissions from f loat ing-roof

petroleum s t o r a g e tanks.

Use of t h e equat ion by t h e r egu la to ry agencies has r e c e n t l y r e su l t ed

i n t h e conclusion t h a t f loa t ing- roof tanks were r e spons ib l e fo r a

s i g n i f i c a n t p o r t i o n of t h e hydrocarbon emission inven to r i e s i n many

A i r Qual i ty Control Regions and t h a t f u r t h e r c o n t r o l was requi red .

Under t h e r e c e n t New Source Review r u l e s , estimates of hydrocarbon

emissions from proposed new tanks have caused cons t ruc t ion of some

of t h e s e tanks t o be blocked.

The opinion has been voiced by some indus t ry s c i e n t i s t s t h a t the

API equat ion does not take i n t o account improvements made i n t h e bas i c

des ign of f loa t ing- roof tanks and roof s e a l s over t h e pas t s e v e r a l

yea r s and t h e r e f o r e a c t u a l emissions could be much less than est imated

by t h e API equat ion. On t h e o t h e r hand, t h e opinion has been voiced

by some regu la to ry o f f i c i a l s t h a t b e t t e r c o n t r o l technology should be

111-1

ENGINEERING-SCIENCE Eq - appl ied t o f u r t h e r reduce f loa t ing- roof tank hydrocarbon emissions

r ega rd le s s of t h e i r r e l a t i v e magnitude.

The Chicago Bridge and I ron Company (CBI), at its P l a i n f i e l d .

I l l i n o i s , r e sea rch f a c i l i t y undertook a test program i n Apr i l 1976 f o r

t h e Standard O i l Company of Ohio (SOHIO) t o examine t h e mechanisms of

hydrocarbon emission l o s s e s from f loat ing-roof tanks and t o ob ta in an

improved hydrocarbon emission loss c a l c u l a t i o n procedure.

equat ion i n API B u l l e t i n 2517, emission es t imates of hydrocarbon l o s s e s

from a typical f loat ing-roof tank were many t i m e s h igher than t h e CBI

estimates.

Using t h e

Laboratory tests were conducted by s t a f f of Chevron U.S.A. Incor-

porated t o eva lua te hydrocarbon l o s s e s from open top con ta ine r s while

varying t h e dead a i r space above t h e l i q u i d hydrocarbons.

conclusions were t h a t very l i t t l e hydrocarbon vapors were l o s t

t o t h e atmosphere when dead space above the l i q u i d approached 30 inches

as commonly found i n f loat ing-roof tank shoe seal designs.

Tenta t ive

The age of t h e emission t e s t d a t a o r i g i n a l l y developed by API

and i t s ques t ionable a p p l i c a t i o n t o today ' s technology, t h e i n i t i a l

r e s u l t s of t h e C B I test program f o r SOHIO, and prel iminary labora tory

r e s u l t s by t h e s t a f f of Chevron U.S.A. Incorporated, a l l suggested

t h e urgent need f o r a re-evaluat ion of hydrocarbon l o s s e s from f l o a t i n g -

roof tanks.

OBJECTIVE

The o r i g i n a l o b j e c t i v e of t h e t e s t i n g program was three-fold:

(a) Determine hydrocarbon emissions from f loat ing-roof tanks a s

a func t ion of seal gap s i z e , vapor pressure , wind v e l o c i t y ,

and o the r important v a r i a b l e s ;

Determine what c o n s t i t u t e d bes t a v a i l a b l e seal technology

and t h e est imated hydrocarbon emissions from use of t h a t

technology; and,

Compare hydrocarbon emissions from e x i s t i n g f loat ing-roof

tanks wi th hydrocarbon emissions est imated t o r e s u l t from

use of bes t a v a i l a b l e s e a l technology.

(b)

(c )

111-2

ENQINEERINQ-SCIENCE IES] - SCOPE -

The f i e l d p o r t i o n of t h i s p r o j e c t w a s l imi t ed t o measuring hydro-

carbon l o s s e s from 1 7 f l o a t i n g roof tanks loca ted in Los Angeles and

San Francisco. No fixed-roof tanks were included i n the s tudy . The

p r o j e c t s i z e and du ra t ion were l i m i t e d by the a v a i l a b i l i t y of tanks.

Most tanks were made a v a i l a b l e f o r t e s t i n g i n August 1976 and were

re turned t o s e r v i c e in January 1977. During t h e t e s t i n g per iod , a

l a y e r of hydrocarbon product was f l o a t e d on water i n most of the tanks

and no product was added.

e s t ima t ing s t and ing l o s s e s ; working l o s s e s were not evaluated.

The eva lua t ion w a s t h e r e f o r e s p e c i f i c f o r

The eva lua t ion of b e s t a v a i l a b l e s e a l technology w a s based on two

surveys.

seals through i n d i v i d u a l companies, both manufacturers and use r s .

second survey addressed s e a l gap s i ze a c t u a l l y measured f o r some 200

tanks i n Southern Ca l i fo rn ia .

The f i r s t survey addressed t h e a v a i l a b i l i t y of commercial

The

Laboratory experiments were conducted by ES t o provide i n s i g h t

i n t o important v a r i a b l e s a f f e c t i n g hydrocarbon evaporat ion rates.

P i l o t scale t e s t s , performed by Chicago Bridge and Iron Company,

were observed and d a t a were eva lua ted f o r p o t e n t i a l input t o t h e

r e l a t io ' n sh ips observed i n t h e ES f i e l d tests.

conducted f irst f o r t h e Standard Oil Company of Ohio (SOHIO) and

la ter f o r t h e Western Oil and Gas Associat ion.

P i l o t s c a l e tests were

Of t h e 17 f u l l - s i z e d tanks made a v a i l a b l e f o r t h i s f i e l d sampling

p r o j e c t , t h r e e of t h e tanks contained crude oil; t h e remaining 1 4 tanks

contained a d i s t i l l a t e product ranging from naphtha t o j e t f u e l

(Table 111-1). A l l t anks m e t the fo l lowing condi t ions :

(a) Roof l e g openings were s e a l e d ;

(b)

(c)

Emergency roof d ra ins were a t least 90 percent covered;

S l o t t e d gauging devices were equipped wi th a f loa t ing -

type Plug;

(d)

(e) A l l t ank gauging or sampling devices were covered, except

Roof guide openings were c losed; and

a t t h e t i m e of sampling.

111-3

ENGINEERING-SCIENCE t%jd -

rl I H H H

w

4 w

al D 3 w .

Y E

e al

5 al U

d $ d

5 al U 01 2 2

5 al Y al 2 d d

I n o u h h I n

d r l d . 0 In m rl

m m m

m U Q m m m u m N r l 4 r l

z z a l d d U

m o u o 4 m m

0 0 rl

111-4

ENQINEERINQ-SCIENCE - METHODOLOGY

Samples of t h e tank con ten t s were c o l l e c t e d at about two-week

i n t e r v a l s and s t o c k depth and temperature were recorded. Sampling

and temperature measurements were genera l ly performed according t o the

methods descr ibed on pages 28 t o 34 o f API Bu l l e t in 2512 ( Ju ly 1957) .

Density of t h e c o l l e c t e d samples was determined by use of a Mettler-Paar

D i g i t a l P rec i s ion Density Meter, Model DMA-50. Hydrocarbon l o s s e s

were then determined by using t h e change of dens i ty method s e t f o r t h

in API B u l l e t i n 2512.

Types of hydrocarbons emi t ted were determined by gas chromato-

graphic ana lyses on samples ob ta ined wi th a bomb in t h e gap between

t h e roof and wa l l .

REPORT ORGANIZATION

The two chap te r s preceding t h i s In t roduc t ion inc lude Summary and

Conclusions of t h e s tudy. The nex t fou r chapters desc r ibe f i e l d

test r e s u l t s , c o r r e l a t i o n s t u d i e s , c a l c u l a t i o n of hydrocarbon l o s s e s ,

and b e s t a v a i l a b l e seal technology.

conclude t h e r e p o r t . A s e p a r a t e r e p o r t of Appendices conta ins a l l

raw test da ta .

Recommendations and re ferences

111-5

._ .. . q

CHAPTER IV

S T A T I C STORAGE HYDROCARBON E M l S S ION MEASUREMENTS

ENGINEERINQ-SCIENCE - CHAPTER IV

STATIC STORAGE HYDROCARBON EMISSION MEASUREMENTS

The basic requirement of this investigation was to determine

the hydrocarbon evaporation loss rate for petroleum stocks during static storage in floating-roof tanks.

escape from floating-roof tanks, the volume flow rate is too low to measure

by any conventional means.

to measure volumetric flow rate and pollutant concentration and then

calculate the pollutant mass flow rate; such procedures cannot be used

for floating-roof tanks.

continuous hydrocarbon monitors downwind from the tank and rely on dis- persion calculations to derive a source emission rate. Major potential

errors associated with assigning dispersion coefficients would probably yield very unreliable estimates of the emission rate.

While hydrocarbon vapors do

The standard source testing method would be

Another measurement procedure would be to locate

Since direct measurement of the loss rate was not possible, ES

utilized an indirect measurement procedure published in an American Petroleum Institute Bulletin,Tentative Methods of Measuring Evaporation Loss from Petroleum Tanks and Transportation Equipment, API Bulletin 2512,

July 1957, pages 28-34, and entitled "Tentative Method for Determining Evaporation Losses by Change in Stock Density." This method relies on

the fact that hydrocarbon compounds that evaporate most quickly from the

tank will be the "light ends" and, therefore, the stock density should

increase after the "light ends" have evaporated.

comprised of hydrocarbon compounds that range from methane to hexane

(C1 - Cg), and petroleum stocks usually contain compounds that range from

C1 - C13. to that of the stock can be estimated by using the density of pentane (C5) and octane (Cg), and is typically 0.85. Thus, evaporation results in less volume of stock in the tank but the volume that does remain has a higher

density than it did initially. The measurement technique, then, is

simply to determine by laboratory analyses the relationship between

density change and the weight change of each petroleum stock under in-

vestigation.

established, stock in the tanks can be sampled to determine initial and

subsequent densities.

a

The vapors are usually

For gasoline, the ratio of the vapor density (in liquid form)

Once this relationship (a density-evaporation curve) is

Increases In stock density are then related to

IV-1

ENQINEERINGSCIENCE - the specific stock's density-evaporation curve to calculate the weight of stock loss.

API Bulletin 2512 also suggested a technique to shorten the weathering time required to see a significant change in stock density and to ensure

that the desired accuracy could be achieved.

utilized for most tanks in this investigation. It involved floating a small layer (two to four feet) of stock on water. This resulted in de-

creasing the volume of the stock that could evaporate and increased the

relative change in density over a short time period. Therefore. a shorter time span was required to measure a given change in stock density.

The suggested technique was

SAMPLE COLLECTION PROCEDURE

The twelve tanks located in Southern California were sampled on a

nominal bi-weekly sampling schedule while the five tanks in Northern

California were sampled once every four to five weeks.

began i n late July 1976 and continued through late December 1976. The field sampling

On each visit, several grab samples of the stock were collected by

water displacement using a sample thief. The samples were obtained by

submerging an 8-ounce narrow neck glass sample bottle filled with water

to the desired position in the stock. verted to allow the stock to displace the water. The sample bottle was

then uprighted, withdrawn from the stock, capped immediately and stored

on ice.

The sample bottle was then in-

In the field sampling procedure, the most significant potential error

was "flashing" of the lighter ends of the petroleum distillate from the

sample bottle before density measurements could be completed. The average

density of the stock had to be measured with an accuracy of 0.00003 g/ml in order to achieve the desired confidence in the loss calculations out- lined in API-2512.

in this study was sufficiently large so as to overshadow all other pos-

sible errors associated with both field and laboratory pfocedures.

potential dictated several procedural modifications as the study progressed.

The initial field sampling procedure used 16-ounce wide-mouth bottles to

obtain the grab samples. Aluminum foil was placed over the mouth of the

jar prior to replacing the cap after the sample had been collected. The

The "flashing" potential of the petroleum stocks used

This

IV-2 . . .

ENGINEERING-SCIENCE - samples were s t o r e d on i c e and de l ive red t o t h e labora tory and then

t r a n s f e r r e d t o o t h e r sample con ta ine r s p r i o r t o dens i ty ana lys i s . These

procedures 'allowed an unacceptable amount of f l a s h i n g and leakage. Sub-

sequent ly , 8-ounce narrow-mouth b o t t l e s were used and samples were not

t r a n s f e r r e d t o o t h e r sample con ta ine r s p r i o r t o the dens i ty determination.

. .

A s a q u a l i t y c o n t r o l check, a l l samples were de l ive red to the

l abora to ry with no i n d i c a t i o n of the s p e c i f i c f loa t ing- roof tank o r sample

p o r t on t h e b o t t l e s . The only n o t a t i o n was a random i d e n t i f i c a t i o n number.

Duplicate samples were given t o t h e labora tory i n random order . A f i e l d

notebook was kept with s p e c i f i c i n s t r u c t i o n s f o r analyses of each sample.

A s igned copy of t h e i n s t r u c t i o n s was given t o the l abora to ry with each

s e t of samples. A s e q u e n t i a l set numbering system was developed t o

ensure a c r o s s check between t h e f i e l d notebook and sample b o t t l e i d e n t i -

f i c a t i o n numbers.

LABORATORY PROCEDURE

The v o l a t i l e na tu re of the petroleum s tocks a l s o requi red changes

i n t h e l abora to ry procedures.

Bingham pycnometer method (ASTM D-1217-54) be u s e d t o determine the

dens i ty of each grab sample and t h a t the densi ty-evaporat ion curve b e

obtained by a bubbler apparatus t h a t increased the rate of evaporat ion.

I n i t i a l l abo ra to ry ana lyses showed t h a t the dens i ty of water and pure

hydrocarbon compounds could be measured t o wi th in t h e prescr ibed

accu rac i e s of 0.00002 g/ml r e p r o d u c i b i l i t y us ing t h e suggested Bingham

pycnometer method.

be obta ined f o r t h e petroleum s t o c k s used i n t h i s s tudy under t h e most

s t r i n g e n t l y c o n t r o l l e d experiments. Seve ra l modi f ica t ions were made t o

t h e pycnometer inc luding using 100 m l wi th a graduated neck r a t h e r than

t h e 25 ml vesse l s . However, no s a t i s f a c t o r y r e s u l t s were obtained.

A p o s s i b l e explana t ion f o r t h e d i f f i c u l t i e s encountered may be t h a t

MI-2512 at tempted t o apply a wider scope of t h e s tock dens i ty change

method than i t was o r i g i n a l l y intended by t h e ASTN committee. Table

I V - 1 shows a quota t ion out of each document t h a t desc r ibes the general

scope of each method. The r e s t r i c t i o n of t h e method i n ASTM D-1217-54

t o "pure hydrocarbons or petroleum d i s t i l l a t e s b o i l i n g between 194 and

The AF'I t e n t a t i v e method suggested t h a t a

However, only 0.00050 g/ml r e p r o d u c i b i l i t y could

IV-3

ENQINEERINQ-SCIENCE - TABLE I V - 1

TENTATIVE METHOD FOR DETERMINING EVAPORATION LOSSES BY CHANGE I N STOCK DENSITY

API BULLETIN 2512 - JULY 1957

PART 11, SECTION D , p. 28

SCOPE

This method is designed t o measure evaporat ion loss by t h e change

i n dens i ty of a s tock dur ing s to rage in any tank or vesse l .

withdrawals can be t o l e r a t e d dur ing a test but any s tock add i t ion voids

t h e t e s t . The method is not s u i t a b l e for pure compounds, very narrow-

b o i l i n g f r a c t i o n s , or for l i q u i d s w i t h a vapor p re s su re g r e a t e r than

0.9 atmospheres a t t e s t condi t ions .

Stock

\ DENSITY AND SPECIFIC GRAVITY OF LIQUIDS

BY BINGHAM PYCNOMETER

ASTM DESIGNATION: D 1217-54

SCOPE

1. (a) This method i s intended f o r t h e measurement of t h e dens i ty

of pu re hydrocarbons or petroleum d i s t i l l a t e s b o i l i n g between 1.94 and

230°F (90 and 1 1 0 0 t h a t can be handled i n a normal fash ion as a l i q u i d

a t t h e s p e c i f i e d t e s t temperatures of 68 and 77'F (20 and 25C).

method w a s developed e s p e c i a l l y f o r t h e r e fe rence f u e l s n-heptane and

iso-octane and is designed t o provide va lues having an accuracy of

0.00003 g p e r m l .

The

(b) The method provides a c a l c u l a t i o n procedure f o r conversion

of d e n s i t y t o s p e c i f i c g r a v i t y .

IV-4

ENGINEERING-SCIENCE (ESI - 2300 F t h a t can be handled i n a normal f a sh ion a s a l i q u i d a t t h e speci-

f i e d t e s t temperatures ," i s not c o n s i s t e n t with the p a r t i c u l a r petroleum

s tocks used i n t h i s s tudy (mostly gaso l ine and crude o i l ) . The example

c a l c u l a t i o n and dens i ty evapora t ion curves presented i n API-2512

imply t h a t t h e method was s u c c e s s f u l l y appl ied a t one t i m e . Therefore ,

r a t h e r than s t a t i n g t h a t t h e technique suggested i n API-2512 is not

c o r r e c t , t h i s experience may only i n d i c a t e t h a t the requi red s k i l l s could

not be gained wi th in a s h o r t t i m e frame t o adequately perform the t e s t s .

Whatever t h e f i n a l r e s o l u t i o n may be , the a p p l i c a t i o n of ASTM D-1217-54

t o gaso l ine and crude o i l should be c a r e f u l l y evaluated before being

used i n any o the r i n v e s t i g a t i o n r e q u i r i n g s t r i n g e n t accuracy.

Other methods of determining dens i ty of gaso l ine and similar s tocks

t o t h e des i r ed accuracy were subsequent ly inves t iga t ed . The Mettler-Paar

D i g i t a l P r e c i s i o n Density Meter, DMA-50, was s e l e c t e d as being t h e most

advantageous method f o r determining t h e petroleum s t o c k dens i ty t o t h e

des i r ed accuracy. The measuring p r i n c i p a l of t h e DMA-50 depends upon

t h e v a r i a t i o n of t h e n a t u r a l frequency of a t u b u l a r o s c i l l a t o r f i l l e d

wi th t h e petroleum s tock . The DMA-50 is c a l i b r a t e d by determining t h e

frequency of o s c i l l a t i o n f o r a i r and b i - d i s t i l l e d water a t a known tem-

pe ra tu re . The known d e n s i t i e s of a i r and water a t t h e test temperature

a r e then used t o determine t h e c a l i b r a t i o n cons tan t f o r t h e tubu la r

o s c i l l a t o r . It is e s s e n t i a l t o know t h e temperature a t which the dens i ty

measurements are obta ined t o w i t h i n +O.0lo C i f f i f t h p l ace d e n s i t i e s

are t o be determined. With t h e DMA-50 t h e r e p r o d u c i b i l i t y has been wi th in

- +0.00002 g/ml f o r a l l petroleum s t o c k s i n t h i s s tudy except crude o i l .

The t h r e e f loat ing-roof tanks t h a t contained crude o i l (Standard 412. 497 ,

and 2140) were dropped from t h e s tudy s i n c e even t h e modified methodology

could not o b t a i n b e t t e r than +O.OOlOO g/ml r e p r o d u c i b i l i t y .

sampling problems a l s o e x i s t e d wi th t h e crude o i l tanks.

F i e ld

I n t h e l abora to ry t h e sample b o t t l e i s r e t r i e v e d from cold s t o r a g e ,

t h e cap is removed and rep laced immediately by a cap with a small hole .

The DMA-50 t e f l o n sample i n l e t tube i s immediately placed through t h e

ho le mid-way i n t o t h e sample b o t t l e . Approximately 10 m l of sample is withdrawn and passed d i r e c t l y through t h e tubu la r o s c i l l a t o r . The

o s c i l l a t o r r e q u i r e s only 0.7 m l of sample, t h e r e f o r e , t h e 10 ml sample

IV-5

ENGINEERING-SCIENCE - ensures t h a t a "mid-stream'' a l i q u o t is obtained.

i s determined and another sample a l i q u o t i s withdrawn a s a check.

more than a +0.00002 frequency d i f f e r e n c e i s observed, another a l i q u o t

i s measured.

The frequency per iod

I f

W i t h d i l i g e n t a t t e n t i o n given t o sample c o l l e c t i o n , handl ing, s to rage ,

instrument c a l i b r a t i o n , and temperature c o n t r o l , i t was observed t h a t

the DMA-50 would accu ra t e ly r epea t the dens i ty measurements t o wi th in

- +0.00002 g/ml. Pure hydrocarbon compounds (hexane and benzene) were

used t o check t h e accuracy of t h e DMA-50 when us ing a i r and water as t h e

c a l i b r a t i o n mediums. The d e n s i t i e s of t h e pure compounds were determined

us ing t h e Bingham pycnometer method f o r f i v e r e p l i c a t e samples.

The DMA-50 c o n s i s t e n t l y gave d e n s i t y measurements t o wi th in +0.00002

g/ml of t h e d e n s i t i e s obtained us ing the 100 m l Bingham pycnometer.

T h e o c i g i n a l API method of determining t h e s t o c k dens i ty by us ing

a pycnometer r e q u i r e s t h a t t h e s t o c k be t r a n s f e r r e d from t h e sample b o t t l e

t o t h e pycnometer wi th a long needle syr inge . Using the DMA-50 t o

determine t h e d e n s i t y , where l i q u i d is syphoned d i r e c t l y from t h e sample

b o t t l e i n t o t h e tubu la r o s c i l l a t o r , reduced t h e poss ib l e e r r o r s s i g n i f i -

can t ly . Simple experiments with t h e DMA-50 ind ica t ed t h a t some gasol ines

could not be t r a n s f e r r e d from the sample b o t t l e t o t h e pycnometer even

a t temperatures as low as 400 F wi thout i nc reas ing the dens i ty by

0.00100 g/ml o r more.

The methodology suggested i n API-2512 t o determine t h e densi ty-

evapora t ion curve involved us ing a bubbler apparatus t o determine the

change i n weight a s soc ia t ed wi th a s p e c i f i c change i n dens i ty .

petroleum s t o c k s used i n t h i s s tudy were s u f f i c i e n t l y v o l a t i l e t o

a l low t h e d e n s i t y evapora t ion curves t o be determined by j u s t allowing

t h e sample b o t t l e cap t o remain o f f f o r s e l e c t e d per iods of t i m e . Af te r

each t i m e interval, the weight change and dens i ty were recorded.

dens i ty evapora t ion procedure i s a d e l i c a t e test i n t h a t small percentage

weight changes must be determined w i t h an a n a l y t i c a l balance. Figure I V - 1

shows a d e n s i t y evapora t ion curve f o r ARC0 134. Four of the twelve

s p a t i a l l y d i s t r i b u t e d samples were used t o determine t h e dens i ty evapora-

t i o n curve. The i n i t i a l d e n s i t i e s of t h e samples were not t h e same

The

The

IV-6

FIGURE I V -

EXAMPLE OENS ARC0 # I 3 4

0 . 7 3 3 0 0

0 . 7 3 2 0 0

0 . 7 3 1 0 0

- E 0 . 7 3 0 0 0 \ M

* I- -

0 . 7 2 8 0 0

0 . 7 2 7 0 0

0 . 7 2 6 0 0 0

TY - EVAPORAT - SEPTEMBER 8

ON CURVE 1976

1 . 0 2 . 0 3.0 4 . 0 5

WEIGHT LOSS, %

ENG I NEER I NG-SC I ENCE , I NC. I V - 7

ENQINEERINQ-SCIENCE - because each had l o s t a d i f f e r e n t amount of hydrocarbon vapor fol lowing

t h e removal of approximately 10 m l of s tock f o r a n a l y s i s with t h e den-

s i t y meter. Even when the i n i t i a l d e n s i t i e s were s i g n i f i c a n t l y d i f f e r e n t ,

no l a r g e change i n t h e s l o p e w a s observed.

dens i ty change inc reases d i r e c t l y wi th inc reas ing weight l o s s , a t l e a s t

i n t h e f i r s t t h r e e percent of the observed weight loss. A least

squares technique w a s used t o determine t h e s lope of the dens i ty evapora-

t i o n curves. The s p e c i f i c s lope f o r each s tock was then used t o de t e r -

mine t h e weight loss assoc ia t ed wi th t h e observed dens i ty change over

t h e "weathering" t i m e span. The appendices l ist a l l l abo ra to ry

r e s u l t s used t o develop each dens i ty evaporat ion curve.

T h i s i nd ica t ed t h a t the

Var i a t ions in s tock dens i ty obta ined from va r ious grab samples

were considered t o r e s u l t from "f lash ing" of l i g h t ends r a t h e r than

from random a n a l y s i s e r r o r . Since f l a s h i n g would r e s u l t i n higher

dens i ty va lues , t h e major random e r r o r i n each sample series should be

t o make c e r t a i n d e n s i t y va lues h ighe r than t h e a c t u a l dens i ty . To o f f -

set t h i s poss ib l e random e r r o r , t h e t h r e e lowest dens i ty measurements

obtained from t h e twelve s p a t i a l samples were always used as t h e b e s t

estimate of t h e average s t o c k dens i ty . The average dens i ty determined

by us ing t h e minimum t h r e e d e n s i t i e s u sua l ly d i f f e r e d from t h e average

of t h e twelve samples by only 0.00001 or 0.00002 g/ml.

ENQINEERINQ-SCIENCE -

EVALUATION OF POTENTIAL STRATIFICATION I N THE TANKS

Severa l samples were c o l l e c t e d t o determine i f t h e r e were any

h o r i z o n t a l or v e r t i c a l s t r a t i f i c a t i o n i n the tanks which would s ig-

n i f i c a n t l y a f f e c t t h e d e s i r e d accuracy of t h e average s tock d e n s i t y .

Whenever poss ib l e , twelve sampling p o r t s were s e l e c t e d f o r each f l o a t -

ing-roof tank s o t h a t each sample r ep resen t s approximately t h e same

volume of s tock . These p o r t s were usua l ly l e g support s l eeves t h a t

extended from one t o two f e e t i n t o t h e s tock . A t least one sample w a s obtained a t mid-depth of t h e petroleum s tock from each p o r t .

Dupl icate samples were o f t e n taken from each po r t t o t e s t r e p e a t a b i l i t y

of t h e sampling method. Since i n i t i a l tests showed no measurable

h o r i z o n t a l s t r a t i f i c a t i o n , twelve mid-point grab samples were obtained

on a l l tanks throughout t h e s tudy per iod . V e r t i c a l grab samples were

obta ined f o r each tank a t least once during the s tudy. These samples

were usua l ly taken out of t h e normal sampling p o r t s r a t h e r than t h e

l e g support s l eeves .

f u r t h e r i n t o t h e s t o c k than the sampling p o r t s l e e v e s and a r e the re fo re

more l i k e l y t o con ta in a s t agnan t l i q u i d .

i nd iv idua l tank , v e r t i c a l s t r a t i f i c a t i o n , i f i t does occur , i s w e l l

below t h e p r e c i s i o n of t h e d e n s i t y measurements and should not a f f e c t

t h e determinat ion of t h e average dens i ty in t h e tank. A s an example,

F igures IV-2 and IV-3 present d a t a c o l l e c t e d for Union 100514 and

Texaco 80210 on September 1 7 , 1976. No hor i zon ta l or v e r t i c a l

s t r a t i f i c a t i o n can be seen t h a t would be l a r g e r than t h e e r r o r s due

t o sampling and ana lys i s .

The l e g suppor t s l e e v e s a r e narrower and extend

Resul t s showed t h a t f o r an

It should be recognized t h a t t h e sampling p o r t s were genera l ly

d i s t r i b u t e d i n t h e tank wi th few l o c a t i o n s be ing nea re r than t h r e e f e e t

t o t h e t ank wal l . One can hypothes ize t h a t i f s t r a t i f i c a t i o n d i d occur

i t would be more pronounced nea r t h e l i q u i d and a i r i n t e r f a c e along

t h e tank w a l l .

t o measuring t h e average s t o c k d e n s i t y i n i t i a l l y and a t t h e end of t h e

"weathering" per iod.

space may a f f e c t hydrocarbon emissions but i t would not s i g n i f i c a n t l y

However, t h e o b j e c t i v e s of t h i s p r o j e c t were l i m i t e d

Any s t r a t i f i c a t i o n t h a t may occur near t h e r i m

FIGURE IV-2

SUMMARY OF TANK T E S T

N

0 4 8 4 8 5

PLANT U N I O N TANK NO. 100514 OATE S E P T . 17. 1976 STOCK DEPTH 74.5' O I A . OF TANK 137'

. 8 2 8 5 +

* A L L NUMBERS I N O I C A T E THE LAST 110 OlGlIS OF T H E 5 t h PLACE OENSITV A S .737 ._ E . m l .

OENSITY MEASUREMENTS NO. OF SAMPLES I7 DEPTH 3' M A X I N U N *.73788 MEAN ,73784 M I N I N U N ,13181 S T D . D E V I A T I O N .00002 AVERAGE OF M I N I M U M THREE ,73782

STOCK TEMPERATURE ( O F )

NO. OF PORTS 13

N l N l N U N SURFACE 71.5

M I N I N U N M I 0 68. I

NAXINUY SURFACE 89.9

M A X I M U M M I D 6 8 . 5

*HAXlNUM NID-DEPTH D E N S I T Y

S P E C I A L D E N S I T Y MEASUREMENTS LOCATION . OENSITY

SP I ' ,13184 S P 3' 13782 S P 5' ,73184 S P 3' ,73784 SP 3' ,73785

ENGINEERING-SCIENCE, INC IV-10

FIGURE 1’4-3

U

SUMMARY OF TANK TEST

PLANT TEX A C O TANK N O . 8 0 2 1 0 DATE S E P T . 1 7 . 1 9 7 6 STOCI( OEPTH 4 5 . I” 011. OF TANK 1 2 0 ’

*ALL NUMBERS l N O l C A T E THE L A S T 110 D I G I T S OF THE 5 t h PLACE D E N S I T Y A S . 6 5 6 _ _ U m l .

DENS I TY MEASUREMENTS NO. OF SAMPLES 15 OEPTH 1 . 5 ’ MAXIMUM *.65606 MEAN .65602 M I N I M U M ,65600 S T O . O E V I A T I O N .00002 AVERAGE OF M I N I M U M THREE , 6 5 6 0 0

S P E C I A L D E N S I T Y MEASUREMENTS LOCAT I O N DEPTH DENS I TY

SP 7 I’ , 6 5 6 0 8 S P 2 2 ” . 6 5 6 0 2 S P 37” , 6 5 6 0 2

STOCK TEMPERATURE ( O F )

NO. OF PORTS 13 MAXIMUM SURFACE 7 4 . I M I N I M U M SURFACE , 7 3 . 0 MAXIMUM MI0 7 0 . 0 M I N I M U M MI0 70 .0

‘MAX IMUM MI 0-OEPTH DENS I T V ,

ENGINEER1 NG-SCI ENCE , INC IV-11

ENQINEERINQ-SCIENCE [Esl-

i n f luence t h e average dens i ty a s determined from the twelve s p a t i a l l y

d i s t r i b u t e d samples.

f i c a t i o n t e s t s which i n d i c a t e t h e t y p i c a l range not iced during t h e

f i e l d t e s t i n g . Because of t h e i n i t i a l problems encountered i n sample

c o l l e c t i o n , handl ing and l abora to ry ana lyses , c e r t a i n portions of t h e

da t a a r e much more r e l i a b l e than o the r s . There a r e two da te s when

s i g n i f i c a n t changes in procedures were incorpora ted which improved

t h e r e l i a b i l i t y of t h e da t a tremendously. On August 20, 1976, a l l

sample t r a n s f e r s were ceased and s t a r t i n g on September 1 7 , 1976,

small-mouth sample b o t t l e s were used f o r a l l sample c o l l e c t i o n

except f o r t h e crude o i l s tock . The u s e of t h e small-mouth b o t t l e s

s i g n i f i c a n t l y reduced t h e v a r i a b i l i t y between d e n s i t i e s measured

throughout t h e tank. Table IV-3 lists some rep resen ta t ive comparisons

between da ta der ived from samples taken a t t h e same depth from t h e

same sampling p o r t . The d e n s i t i e s determined us ing t h e wide-mouth

b o t t l e s were always s i g n i f i c a n t l y h i g h e r than the d e n s i t i e s ob ta ined-

us ing small-mouth b o t t l e s , which i n d i c a t e d t h a t more l i g h t ends

evaporated wi th t h e use of wide-mouth bot. t les. For t h i s reason, no

d e n s i t y d a t a c o l l e c t e d p r i o r t o September 1 7 , 1976 were used t o esti-

mate t h e abso lu te change in average s t o c k dens i ty during t h e s tudy per iod .

Table I V - 2 l ists s e v e r a l examples of t h e s t ra t i -

TABLE IV-2

EXAMPLE DEPTH STRATIFICATION RESULTS

S p a t i a l Depth Var ia t ion Var i a t ion Surface Minus Bottom

Tank Date (10-5 g/ml) (10-5 g/ml)

Union 100514 9/17/76 5 0

Texaco 80210 9/17/76 6 +6

Exxon 1757 9/24/76 14 +8

Gulf 80005 10/28/76 11 +14

ARC0 134 10/28/76 4 +8

Union 100514 11/11/76 20 +1

Union 330 12/03/76 3 +1

I V - 1 2

. . I

ENQlNEERlNGSClENCE /ES( - TABLE IV-3

A COMPARISON OF DENSITY MEASUREMENTS WIDE MOUTH BOTTLES VERSUS

SMALL MOUTH BOTTLES*

Density (g/ml) Maximum Ap Tank Date Wide Smal l ( g / W

She l l 47

Standard 9464

Standard 192

Union 5

10/13/76

10/12/76

10/12/76

12/14/76

0.73917

0.73911

0.73907

0.77721

0.77701

0.77693

0.69304

0.69287

0.69281

0.73865

0.73865

0.73887

0.73806

0.77693

0.77604

0.77680

0.69273

0.73857

0.73858

.00031

,00041

.00031

.00008

* A l l samples f o r each t ank were c o l l e c t e d from one p a r t i c u l a r sampling p o r t .

1.V-13

ENQINEERING-SCIENCE IES] - EFFECTS OF WATER, TEMPERATURE AND FLASHING TIME

Simple l abora to ry experiments were conducted to eva lua te the in f luence

of c e r t a i n v a r i a b l e s on t h e dens i ty measurements.

t h e tanks w a s quest ioned a s t o t h e poss ib l e change i t would have on t h e

dens i ty measurements. Water was purposely l e f t in a few sample b o t t l e s

taken from a tank.

water and gaso l ine .

f o r a t l e a s t t h r e e days. No no t i ceab le v a r i a t i o n in the dens i ty was ob-

served between those samples conta in ing water and those not conta in ing

water.

dens i ty v a r i a t i o n due t o a i r d i s so lv ing i n t o t h e s tock . The dens i ty of

hexane w a s determined a t room temperature , then t h e hexane w a s heated t o

90° F and t h e dens i ty w a s t h e same as a t room temperature.

hexane was then cooled from 90° F t o 400 F and l e t s t and f o r t h r e e hours.

No measurable change in dens i ty w a s observed.

approximately 15 seconds l a p s e from the t i m e a sample b o t t l e is withdrawn

from t h e s t o c k t o t h e t i m e t h e cap is placed on t h e b o t t l e . Ques t ions

were r a i s e d as t o the amount of f l a s h i n g t h a t could occur during t h e

time period. Gasol ine was p laced in a sample b o t t l e a t room temperature.

The cap was removed and t h e d e n s i t y w a s determined. The cap w a s l e f t o f f

f o r 15 seconds and no change in d e n s i t y w a s no t iced .

l e f t of f f o r 30 seconds and aga in no change i n dens i ty was not iced . It

w a s concluded t h a t t h e f l a s h i n g t h a t occurs dur ing t h e t i m e i t t akes t o

p l ace t h e cap o n , t h e sample b o t t l e does not e f f e c t t h e dens i ty wi th in

the accuracy range of t h e d e n s i t y meter.

The use of water in

The samples were shaken t o ensure more con tac t of t h e

The dens i ty of each was determined a f t e r being s t o r e d

Temperature of t h e s tock has been r a i s e d as a p o s s i b l e reason f o r

The sample

During f i e l d sampling,

The cap was then

IV-14

ENQINEERINQ-SCIENCE - EXCLUSION OF CERTAIN TANKS

Three crude o i l tanks (Standard 412, 497, and 2140) a r e shown in

Tables IV-12. 13, and 1 4 but a r e not considered f u r t h e r in t h i s study

s i n c e very u n r e l i a b l e r e s u l t s were obtained. Incons i s t en t r e s u l t s f o r

t h e crude o i l may have been caused by handling problems in the labora tory

procedure or by t h e f a c t t h a t a r e s idue w a s formed in each tank between

the o i l and water i n t e r f a c e . Moreover, as s tock temperature decreased

during t h e s tudy , t h e crude oil became very v iscous , and s t r a t i f i c a t i o n

may have occurred. Analyses of t h e hydrocarbon l o s s e s from these tanks

showed mostly methane and ethane so t h e i r importance may be less s i g n i f -

i c a n t from an a i r p o l l u t i o n p o i n t of view than the gaso l ine s to rage tanks.

Union 5 (Table IV-18) is a l s o n o t descr ibed f u r t h e r s i n c e a l a r g e

random v a r i a t i o n of s tock dens i ty was not iced from one sampling po r t

t o another . Density da t a from Union 5 were i n c o n s i s t e n t and c o l l e c t e d

samples d i d not adequately r ep resen t the s tock s i n c e t h e l e g support

sleeves reached nea r ly t o the g a s o l i n e and water i n t e r f a c e .

DENSITY MEASUREMENTS

The a c t u a l f i e l d no te s h e e t s and labora tory a n a l y s i s s h e e t s are

l i s t e d in t h e appendices of Volume 2.

(F ie ld Measurement Summary), a t t h e end of t h i s Chapter, summarize the

p e r t i n e n t d a t a f o r each f loa t ing- roof tank. The d e n s i t i e s l i s t e d in

t h e F ie ld Measurement Summary s h e e t s a r e f o r those samples taken

h o r i z o n t a l l y throughout t h e tank a t mid-depth of the s tock . The column

l abe led "MIN. 3 AVERAGE" r e p r e s e n t s t h e b e s t estimate of t h e t r u e average

s tock dens i ty .

Tables I V - 4 through IV-20

Figures IV-4 through IV-7 are g raph ica l r ep resen ta t ions of t h e

dens i ty change dur ing t h e "weathering" per iod .

a smooth curve o r s t r a i g h t l i n e t h a t could be drawn between t h e d a t a

po in t s i n FiguresIV-4 through IV-7 tends t o be a measure of t h e

day-to-day sampling and a n a l y s i s e r r o r r a t h e r than a r ap id change in

dens i ty .

ending October 29, 1976 shows t h a t t h e measured d e n s i t i e s of six s tocks

decreased. These tanks were: ARC0 134, Gulf 80005, S h e l l 47 and S h e l l

48, and Standard 192 and Standard 9464. A decrease in s tock dens i ty is

The s c a t t e r about

A c l o s e examination f o r t h e d e n s i t i e s measured f o r t h e week

IV-15

F.IGURE IV-'

CHANGE I N STOCK DENSITY WITH TIME

- E k ul

I 0 - r I-

v) z Y 0

Y 0 0 I- v)

-

73600

13500

73400

73300

76500

76400

1 6 3 0 0

1 6 2 0 0

D A Y ~ O N T H

ENGINEERING-SCIENCE, INC IV-16

FIGURE I V -

69100

CHANGE I N STOCK OENSITY WITH T I M E

S T 0 192 I

74000

73900

73800

73100

74100

74000

73900

13800

ENGINEERING-SCIENCE, I N C . IV-17

F I G U R E I V -

CHANGE I N STOCK D E N S I T Y WITH T I M E

74100

74000

73900

73800

-̂I T X 20010

OAY/MONTH

ENGINEERING-SCIENCE, I N C . IV-18

FIGURE IV-7

In I 0 -

CHANGE I N STOCK DENSITY WITH T I M E

DAY/MONTH

DAY/’MONTH

ENGINEER ING-SC I ENCE, INC IV-19

ENGINEERING-SCIENCE - contrary to the fact that as evaporation occurs, the density increases.

Subsequent data points for each of the six tanks suggest that the

above mentioned densities were erroneous. The six tanks were sampled,

two tanks each day, on October 26, 27 , and 28. Laboratory analyses were

performed on October 28 or November 1. No other tanks were sampled

between October 26 and 28 nor were any other samples analyzed in the

laboratory on October 28 o r November 1. A s previously discussed, all

expected errors result in measured densities which are higher than the

true densities. Therefore, the six data points mentioned above should

be treated as suspect.

The emission rate estimate is based upon the density-evaporation curve and the change in density over the weathering period. The change

in density was estimated using linear regression and an examination of

the beginning and end point densities.

have to be made between the two techniques. In each of those cases, the higher of the two density changes was used to estimate the emission rate. The intent was to estimate the upper limit emission rate rather

than a lower one. Table IV-21 lists the number of days estimated change

in density, slope of the density evaporation curve, and estimated

emissions f o r each of the thirteen tanks.

In some cases, a choice would

IV-20

, . t

ENGINEERING-SCIENCE ES "1 ',

..

Y =I U

-2-

L 0

Y 0 > c

I I'J- 2 1 I

ENGINEERING-SCIENCE -

..

A Y Y)

Y Y Y O O Y O

Y e c = w o w a - 2 - O W O Y Z S U E

0 Y 0

p: 0 2 0 v

!

I

ENGINEERING-SCIENCE rEs] -

IV-23

ENGINEERING-SCIENCE -

!

' I

!

A / . . c1

c L 1

TV-24

ENGINEERING-SCIENCE -

-2-

zv-2s

ENGINEERING-SCIENCE P I -

;

L -

m VI-

IV-26

ENQINEERINQ-SCIENCE -

!

L

111-27

ENQINEERINQ-SCIENCE -

..

IV-28

ENGINEERING-SCIENCE E s "1

-2-

I IV-29

ENGINEERING-SCIENCE -

- . Y * c E

Q . 4

-Z-

YI

IV-30

1 ENGINEERING-SCIENCE -

, i L

I- f

I ! e 5 0 L

I- f

! I

L D

Y 0 - I-

IV-31

ENGINEERING-SCIENCE -

E e Y

6

r : E 5 5

:

c

a

.n 3

I\'-32

h

ENQINEERING-SCIENCE -

. .

5

!

IV-33

ENGINEERING-SCIENCE -

i I

-2-

e I c P -

IV-74

ENGINEERING-SCIENCE -

-2-

!

L

IV-35

ENGINEERING-SCIENCE -

-2-

c )1

c : - Y = a . 0 .

W

IV-36

ENQINEERINQ-SCIENCE IES] -

-2-

0

P

YI

z z ‘ o f

IV-37

ENGINEERING-SCIENCE IES] -

m l . o m 0 m l . m r D m m u r D r . \ o m m u m m u m m 3 U N

m m

m U m

0 rD U

U 3

m N 0 0 m m

m N 0 m

m U rl

N ro

l. l.

N rD

U m

u m m m

0 m rD o\

U co o\ m 0 m

0 N x 0 0 U d d e 3

N

rl

71 u m 71 e m U cn

m U rD U

71 U m 0

U cn

m

2

0 rl 0 z 0 U

$ c

2 N 0 oc 0 U m w 3

U 3 m 0 0 rl

c 0 d 0 3

m 0 0 0 W

W 3

CI a

0 rn m

g d e 3

m N U G 0 d 4

CHAPTER V

CORRELAT I ON OF EM1 SS IONS W I T H MEASURED PARAMETERS

i

CHAPTER V

CORRELATION OF EMISSIONS WITH MFASUKED PARAMETERS -

GENERAL

The hydrocarbon emissions were measured f o r t h i r t e e n tanks a s

discussed in Chapter I V . In t h i s Chapter, t h e measured emissions w i l l

be compared t o t h e W I equation f o r es t imat ing loss r a t e s and a l s o t o

c e r t a i n parameters t h a t were measured during the s tudy per iod. All p e r t i n e n t parameters needed f o r t h e APT equat jon were measured. How-

ever , i t is not reasonable t o assume t h a t all parameters t h a t may

a f f e c t the a c t u a l emissions were measured or ca l cu la t ed . For in s t ance ,

on shoe s e a l s , t h e p i l o t p l a n t s t u d i e s performed by C S I have j u s t

r e c e n t l y shown t h a t t h e condi t ion of t h e f a b r i c apron above the vapor

space between the shoe and t h e pontoon, has the g r e a t e s t i n f luence

on emission r a t e s f o r t h e s c e n a r i o s they have considered. This

parameter was not considered a t t h e beginning of t h i s s tudy and w a s

never measured o t h e r than a b r i e f v i s u a l inspec t ion . However, i t i s

reasonable t o assume t h a t t h e f a b r i c on newer seals may be in b e t t e r

r o n d i t i o n than the f a b r i c on o l d e r seals. The age of t h e shoe seals

a r e known and w i l l be d iscussed as a p o s s i b l e i n d i c a t o r of o t h e r non-

measured parameters.

It should be s t r e s s e d t h a t w i t h only t h i r t e e n tanks a s b a s i c da t a

p o i n t s , and cons ider ing t h e l a r g e number of poss ib l e parameters t h a t

may a f f e c t t h e emission rates, only gene ra l conclusions regarding

t r ends a r e poss ib l e .

Table V-1 l ists t h e tanks and summarizes what are thought t o be

t h e most important of t h e measured parameters. Most of t h e tanks

contained gaso l ine wi th a Reid vapor p r e s s u r e of 9 p s i . The tank

d iameters ranged from 55 t o 153 f e e t . Most tanks had shoe s e a l s . The

measured s e a l gap space between t h e tank s h e l l and s e a l ranged from

zero t o 8.6 square f e e t . The number of gaps ranged f r m zero t o 72.

The shoe s e a l s were i n s t a l l e d between 1951 and 1968 and the tube s e a l s

were i n s t a l l e d between 1962 and 1976. The measured emissions ranged

from nea r ly ze ro t o 265 pounds pe r day.

v- 1

ENQINEERINQ-SCIENCE (ESI -

m o m o o o o o m o p . m o m o ~ m o m m r l u h o . . . . G d d a i o r l m o

U h 0 0 0 N U 0 0 0 o u o o m . . . . . o o o o \ o

v-2

ENGlNEERlNQ-SCIENCE - COMPARISONS WITH API EQUATION

The American Petroleum Institute in their API Bulletin on Evaporation

Loss Prom Floating-Roof Tanks, API Bulletin 2517, February 1962, outlines a

suggested procedure for estimating standing-storage loss from floating-roof

tanks. The recommended equation is

where: L

Kt - a tank-type factor which changes as follows: - standing-storage evaporation loss, in barrels per year. Y

Kt - 0.045 for welded tank with pan or pontoon roof, single or

Kt - 0.11 for riveted tank with pontoon roof, double seal. Kt - 0.13 for riveted tank with pontoon roof, single eeal.. Kt * 0.13 for riveted tank with pan roof, double seal. Kt - 0.14 for riveted tank with pan roof, single seal.

double seal.

D - tank diameter, in feet [For tanks 150 ft or less in diameter, use D1*5; for tanks larger than 150 ft in diameter, use 1501'5 (=)I D

P - true vapor pressure of the stock at its average storage temperature, in pounds per square inch absolute.

V, - average wind velocity, in miles per hour. K - a recommended eeal factor: s

Ks - 1.00 for tight-fitting seals. Ka - 1.33 for loose-fitting seals.

K - a recommended factor distinguishing between gasoline and crude oil storage:

Kc - 1.00 for gasoline. K - 0.75 for crude oil. C

R - a recommended paint factor for color of shell and roof: P

K - 1.00 for light gray or aluminum. K * 0.90 for white. P

P

v-3

ENQINEERINQ-SCIENCE - The above equat ion was app l i ed t o a l l t h i r t e e n tanks i n t h i s s tudy.

F igure V-1 shows t h e c o r r e l a t i o n of measured emissions and those calcu-

l a t e d us ing t h e API equat ion. Table V-2 lists t h e two emission rates f o r

each tank. Nine tanks e m i t less than the API es t imated emissions and

fou r tanks emit more than t h e API estimated emissions. The emission esti-

mate f o r Texaco 20010 is very c l o s e t o t h e observed emission r a t e .

However, o v e r a l l t h e observed emission rates were approximately 58 percent

of t h e ca l cu la t ed emission r a t e s .

equa t ion Overs ta tes t h e magnitude of t h e emissions and does not adequately

r ep resen t t h e in f luence of s p e c i f i c parameters.

It would appear t h a t the API 2517

TABLE V-2

API 2517 CALCULATED VERSUS OBSERVED HYDROCARBON EMISSIONS

Hydrocarbon Emission Rate (bbls /year ) API 2517

Tank Calcula ted Observed

ARC0 510 55

Exxon 30 87

Gulf 131 - 0

Lion 128 175

Mobil 49 60

S h e l l 47 237 55

S h e l l 48 237 57

Standard 192 362 445

Standard 9464 35 56

T W c o 20010 34 33

TexAco 80210 145 43

Union 330 132 84

Union 100514 433 286

v- 4

\

~~

F I G U R E V-1

- ENGINEER1 NG-SCIENCE, IN[ v- 5

\

\ 0 0 -

0 0 m

0

ENQINEERINQ-SCIENCE IES] - CORRELATION WITH MEASURED PARAMETERS

Ffgures V-2 through V-4 show the c o r r e l a t i o n of measured emission and

c e r t a i n measured parameters t h a t a r e thought t o be important t o t h e

product ion of hydrocarbon emissions. Because of the many parameters

involved, cau t ion should b e used i n eva lua t ing only one parameter a t a

time; however, i t i s i n t e r e s t i n g t o ga in i n s i g h t i n t o the poss ib l e e f f e c t

one v a r i a b l e may have on t h e emission r a t e s . As shown, t h r e e tanks have

emission r a t e s much h igher than the o t h e r s . Standard 192, Union 100514

and Lion 428 emission r a t e s were from two t o f i v e times h igher than a l l

t h e o t h e r tanks. Measured parameters such as wind speed, Reid vapor

p re s su re , s tock temperature of s i z e of t h e seal gaps ind iv idua l ly do not

appear t o t o t a l l y account f o r t h e t h r e e h igher emissions. As mentioned

e a r l i e r , t h e r ecen t C B I s t u d i e s have shown t h a t the t i g h t n e s s of the

vapor space under t h e f a b r i c s e a l on m e t a l l i c shoe seals is a very important

v a r i a b l e .

A pres su re test may b e requi red t o make t h e f i n a l judgment concerning t h i s

f a c t o r f o r Standard 192. It is i n t e r e s t i n g t o no te t h a t t h e primary

seal on Standard 192 was i n s t a l l e d i n 1951 and r ep resen t s t h e o l d e s t

primary s e a l i n t h i s s tudy. The next o l d e s t seal i s Union 100514 (1952).

Figure V-5 shows t h e measured emission r a t e s ve r sus the year the primary

seal w a s i n s t a l l e d on each tank. A v i s u a l i n spec t ion w a s made of the

f a b r i c apron of t h e o t h e r shoe seals and a l l had apparent ho le s i n t b e f a b r i c .

No tears were v i s u a l l y observed f o r t h e Standard 192 tank.

The C B I p i l o t tank s tudy o f f e r s an e x c e l l e n t oppor tuni ty t o eva lua te

t h e in f luence of one v a r i a b l e wh i l e holding a l l o t h e r v a r i a b l e s cons tan t .

The p i l o t scale s t u d i e s on a twenty-foot tank have shown t h a t the e m i s -

sions vary almost d i r e c t l y wi th vapor p re s su re which seems t o agree wi th

t r a d i t i o n a l t h e o r e t i c a l cons ide ra t ions . However, s e v e r a l tests would

i n d i c a t e t h a t c e r t a i n v a r i a b l e s t h a t have previous ly been considered

t o d i r e c t l y impact t h e emissions may not be very important.

The d a t a on shoe seals, f o r example, show t h a t when t h e gap area is

increased by a f a c t o r of two the emission a c t u a l l y tended t o remain the

same or decrease .

of two gaps each 0.5 inch wide and two f e e t long.

was comprised of two gaps each 1 .0 inch wide and two f e e t long.

I n t h e above example, t h e f i r s t gap area was comprised

The second gap a r e a

The r i m

V- 6

FIGURE V-

HYDROCARBON E M I S S I O N RATE AS A FUNCTION - OF R E I D VAPOR PRESSURE AND TANK DIAMETER

2 7 5

2 5 0

225

z o o

175

150

125

IO0

7 5

5 0

25

0 0 2 6 8 10 12 I4 16 I8

R E I D VAPOR PRESSURE, p s i

0 20 4 0 . 60 80 100 I20 I40 160 180

TANK D I A M E T E R , f t .

ENGINEER ING-SCI ENCE, INC v-7

FIGURE V-

75

50

25

0

HYDROCARBON E M I S S I O N RATE AS A FUNCTION OF_ AVERAGE WIND SPEEO AND AVERAGE AMBIENT TEMPERATURE

I I U M 330

UOBl L 7 E X X O N w

STO 5464 A R C 0

smL 18 -01x 20010

GULF 18 14 16 0 2 6 8 I O 12

AVERAGE WIND SPEED, mph

2 7 5

250

2 2 5

200

175

150

125

IO0

7 5

50

2 5

0 40 4 5 50 5 5 6 0 6 5 7 0 75 80 6 5

AVERAGE AMBIENT TEMPERATURE O F

ENGINEERING-SCIENCE. I N C . v-8

FIGURE V-4

z 0

4 0 0 = =- I

m a

n

W I- U a

z 0 r n a U 0 0

0 =- I

a

HYDROCARBON E M I S S I O N RATE AS A FUNCTION - OF SEAL GAP AREA AND NUMBER OF GAPS

0 1 . 0 2 . 0 3 . 0 4 . 0 5 . 0 0 . 0 7 . 0 8 . 0 9

SEAL GAP AREA, f t 2

0 1 . 0 2 . 0 3 . 0 4 . 0 5 . 0 0 . 0 7 . 0 8 . 0 9

SEAL GAP AREA, f t 2

v-9 ENGINEERING-SCIENCE, I N C .

FIGURE V-'

0 0 0

0 0 0 0 0 y1 y1 k7 0

n n z - d e P / s q l ' 3 1 V M NOISSIW3 NOBYV30MOAH

0

ENGINEER ING-SCIENCE. INC v-10

. . I

ENQINEERINQ-SCIENCE - space temperature w a s heated t o ZOO F above the bulk l i q u i d temperature

and i n s t e a d of t h e emissions inc reas ing a s one might expect due t o the

inc rease i n vapor p r e s s u r e , the emissions tended to decrease or remain

t h e same. Perhaps t h i s i n d i c a t e s t h a t the "noise" l e v e l of t h e t e s t

procedure is below the range of t hose two va r i ab le s . A s i g n i f i c a n t

i nc rease i n emissions was not iced when the gap a r e a was increased by a

f a c t o r of e i g h t . The gap was 0 .5 inch w i d e and t h i r t y f e e t long (halfway

around t h e circumference) i n the above example f o r t h e shoe seal.

Secondary s e a l s tended t o decrease the emissions f o r the C B I tests. A l l

of t h e pre l iminary p i l o t t e s t s on shoe s e a l s tended t o show t h a t the

emissions increased wi th t h e amount of d i l u t i o n air blown through t h e

test chamber. This would i n d i c a t e t h a t wind speed c e r t a i n l y a f f e c t s the

emissions; however, t h e exac t r e l a t i o n s h i p i s d i f f i c u l t t o determine due

t o r e l a t i n g t h e d i l u t i o n a i r f low rate t o wind speed. S imi la r r e s u l t s

were found wi th tube s e a l s . In f a c t , i t would appear t h a t a t i g h t

f i t t i n g (no gap) tube s e a l would have h igher emissions than a t i g h t

f i t t i n g (no gap) shoe s e a l . Furthermore, t h e same gap space area f o r a

tube seal would y i e l d h igher emissions than f o r the same gap space a r e a

f o r a shoe s e a l .

While t h e C B I d a t a are much more u s e f u l than t h e f i e l d sampling

d a t a t o show dependency upon one v a r i a b l e , i t i s of in te res t t o s e e i f

similar t r ends e x i s t i n t h e f i e l d da t a . I f a l l of t h e t h i r t e e n tanks

are cons idered , even those with o l d seals and obvious f a b r i c epron problems,

it would appear t h a t emission may be inf luenced by Reid vapor p re s su re ,

s t o c k temperature , wind speed and t o some e x t e n t , gap s i z e . However, if

the t h r e e o l d e r tanks , two of which have obvious f a b r i c ho le s are

d e l e t e d , then no conclusions can r e a l l y be drawn t h a t would imply that

t h e emissions a r e r e l a t e d s u b s t a n t i a l l y t o Reid vapor p re s su re , s tock

temperature , wind speed o r gap s i z e .

exist between the f i e l d experiment and t h e p i l o t p l a n t test.

Lion, Standard 192, and Union 100514 tanks a r e neglec ted , then the two

h ighes t emitters (of t h e t e n remaining tanks) had tube seals on welded

tanks wi th no n o t i c e a b l e gaps.

Gul f 80005. I t has t h e newest shoe s e a l (1968) but had 23 gaps t h a t

I t i s apparent t h a t one p a r a l l e l does

If t h e

The lowest emission rate w a s found on

t o t a l e d 0.56 f t L and observable openings In the f a b r i c apron. However,

v-11

P

EN~INCERIN~.S~IENCE - t h e vapor p re s su re f o r Gulf 80005 was a l s o one of t h e lowest i n the

study. - I n gene ra l , i t does appear t h a t the f i e l d d a t a and the p i l o t test

da t a do agree very w e l l i n t h a t t i g h t f i t t i n g shoe s e a l s and t i g h t f i t t i n g

tube seals would l o s e approximately t h e same amount of s tock . Fu r the r ,

i t would appear t h a t tube seals, whi le decreas ing the gap s i z e s i g n i f i -

can t ly under most cond i t ions , do not o f f e r a n advantage over shoe seals

a s a p o l l u t i o n c o n t r o l device.

The s imple l abora to ry experiments conducted dur ing t h i s s tudy tend

t o confirm t h a t tube s e a l s should be more s u s c e p t i b l e t o wind condi t ions

than shoe seals. C B I a l s o found t h a t wind is more of a f a c t o r f o r tube

seals.

The C B I p i l o t t e s t s tend t o show t h a t emission r a t e has a s t r o n g

dependency on wind speed and t r u e vapor pressure i f a l l o t h e r v a r i a b l e s

a r e he ld cons tan t . I f t h i s is t r u e , one would expect t h a t as t h e ambient

wind speed decreased and t h e s t o c k temperature decreased ( a f f e c t i n g t r u e

vapor pressure) then t h e emission rate would a l s o decrease f o r a pa r t i cu -

l a r tank. If t h e emission rate decreased then t h e r a t e a t which t h e

average s tock dens i ty inc reases would not remain cons tan t . For a l l tanks

in t h i s s tudy , t h e wind speed and s tock temperature have decreased during

t h e s tudy per iod b u t t h e rates at which the d e n s i t i e s increased have

appeared t o remain cons tan t .

F igure V-6 is a p l o t f o r Standard 192 ( t h e h ighes t e m i t t e r ) comparing

t h e changes in dens i ty wi th t h e observed s tock temperatures , and ambient

wind speed.

of 8.9.mph a t t h e end of September t o only 4.5 mph a t t h e end of

December.

t h e rate of i n c r e a s e in d e n s i t y seems t o have remained cons tan t .

As s h a m , t h e wind speed decreased from a two week average

The s tock temperature decreased from 690 F t o 60' F. However,

I t should be s t r e s s e d t h a t the in f luence t h a t only one v a r i a b l e has

on the emission rate of a f loa t ing- roof tank i s a t bes t , d i f f i c u l t t o

a s c e r t a i n . A c o n t r o l s t r a t e g y t o reduce t h e emissions, if requ i red , should

cons ider parameters o t h e r than seal gap s i z e .

seventeen tanks i n t h e s tudy and shows t h e i r compliance s t a t u s wi th Rule

463 as of January 15, 1975. Five tanka are in compliance and twelve are

Table V-3 lists a l l of t h e

v-12

FIGURE V-l

- STANDARD 192 COMPARISON OF W I N D S P E E D AND STOCK TEMPERATURE

TO CHANGE I N D E N S I T Y

6 . 0

5.0

4 . 0

4 :I 0 . 6 9 4 0 0

0 . 8 9 3 0 0

0 . 6 9 2 0 0 1 7 24 1 6 15 22 29 5 12 1 9 26 3 1 0 17 24 3 1

S E P T E M B E R O C T O B E R N O V E M B E R D E C E M B E R

0 A Y /MONTH

ENGINEERING-SCIENCE. I N C . v- 13

ENGINEERING-SCIENCE -

g d U LE 0 d w d U

5 H

Y

H

5

3

v l v l o o o m m VI O O N h v l v l v l h r l O ~ O O h O v l O

4 4 4 4 . . . . . . . . . . .

o o m ~ a m m m v l o a o o ~ o h a m Nr. N N U r l N r . N d 4

4

m vl 4

h w v l 0 0 v l 0 0 u 0 0 v l a 0 m r l h m m m 4 0 ~ m m ~ m m m ~ rl 4 4 4 4 4 4

rl rl

2 3

5

Y

U

al

U U m U C 0 0

U 0 C a d 5

4 m' al m al

5 3

g

d c 5 00

4 m al

$ U

.rl 'El

al c

rn

a. m U

5 z Y a al I,

M al

9 C

0 .rl c h

.. al U 0 z I

v-14

ENQINEERINQ-SCIENCE - not. Only those tanks with tube s e a l s m e t the gap s i z e rule . Similar

tanks that did not meet the ru le had emission rates lower than the

f i v e i n compliance.

V-15

CHAPTER V I

FURTHER OlSCUSSlON OF S I GNI F ICANT PARAMETERS

ENQINEERINQ-SCIENCE -

CHAPTER V I

FURTHER DISCUSSION OF SIGNIFICANT PARAMETERS

Table V I - 1 p r e sen t s f o r comparison a l i s t of v a r i a b l e s o r f a c t o r s

included in t h e API-2517 equat ion f o r emission l o s s from f loat ing-roof

tanks and in t h e ES c o r r e l a t i o n stqidy of f i e l d t e s t da t a . The API

equat ion is defined i n Chapter V .

t h e ES f i e l d s tudy, and t h e API s tudy d i d not inc lude double-deck type

r o o f s . Therefore , i t i s not known i f any e f f e c t on emission loss can

be a t t r i b u t e d t o d i f f e r e n c e s between pan, pontoon, o r double-deck

roofs .

There were no pan-type roofs i n

In t h e ES s tudy , tank cons t ruc t ion , roof type , and s e a l t y p e were

represented a s c a t e g o r i e s f o r c o r r e l a t i o n purposes. In t h e API

equat ion , t h e e f f e c t s of t hese f a c t o r s were combined i n a v a r i a b l e ,

k t , f o r which numerical magnitudes were given depending on t h e com-

b i n a t i o n of tank, roo f , and s e a l type f o r a s p e c i f i c tank. Also,

t h e API p r e d i c t i o n equat ion does not d i s t i n g u i s h tube or shoe type

s e a l s , and does not e x p l i c i t l y t r e a t s e a l gap a r e a o r number of s e a l

gaps a s an emission r a t e v a r i a b l e . It does, however, i m p l i c i t l y in-

f o r o ld o r new s e a l s , and i n t h e c lude t h e e f f e c t in a cons t an t ,

k v a r i a b l e . ks

t

One of t h e main d i f f e r e n c e s , then, between t h e API p r e d i c t i o n

equat ion and t h e ES f i e l d experiment d a t a c o r r e l a t i o n s t u d i e s i s t h a t

ES examined t h e new f i e l d measured d a t a t o eva lua te any d i r e c t quant i -

t a t i v e r e l a t i o n s h i p between seal gap numbers or area and emission loss

independent of type of roof and s e a l , whereas the AF'I equat ion assumes

i m p l i c i t l y t h a t gap a r e a s are always more f o r o ld seals, o r r i v e t e d

t a n k s with a s i n g l e seal on i ts roof .

a l s o i m p l i c i t l y assumes t h a t t h e emission loss from any p a r t i c u l a r

t a n k / r o o f / s e a l type combination is independent of s e a l gap area s i n c e

t h e va lues of k

seal combination are s p e c i f i e d .

However, t h e AF'I equat ion

and kt a r e f i x e d once t h e age f a c t o r and tank/ roof / S

V I - 1

ENGINEERING-SCIENCE - TABLE VI-1

COMPARISON OF ENGINEERING-SCIENCE FIELD TEST VARIABLES TO API 2517 EQUATION VARIABLE

Engineering-Science Field Test API 2517 Study Emission Loss Correlating Emission Loss Equation

Variable or Factor* Variable or Factor

1. Roof Type (Pontoon - double deck) 2. Roof Color

3. Tank Diameter, D (ft) 4.,Seal Type ( shoe, tube orother

5 . Seal Gap Area (ft )

6. Number of Seal Gaps 7 . Reid Vapor Pressure ,RVP

2

or True Vapor Pressure (psia)

8. ASTM Dist. Curce Slope 9. Average Stock Temperature (OR) 10. Emission Rate (lb/day, bbls / year) 11. Average Ambient Temperature (OR ) 12. Average Maximum Hour Temperature (OR) 13. Average Day Temperature Range, AT (OR)

14. Average Ambient Pressure (in. Hg) 15. Average Wind Speed (mph) 16. Average Maximum Hour Wind Speed (mph) 17. Average % Cloud Cover

** Tank Type ( Welded - Riveted)

k ,function (pan - pontoon) Paint Factor, K t

P D (It)

k function (single seal -

k function (new - old) double seal) t’

S ’ - True Vapor Press., pv(psia) k function (gasoline -

crude oil) C’

- L (bbls / year) Y -

- 14.7 (psia)

V (mph) W - -

k- function (welded - riveted) t’

* As numbered on summary data sheets. ** Not numbered on data sheets.

VI-2

~

ENQINEERING-SCIENCE - Therefore , a l though t h e measured emission loss f o r t h e 13 s to rage

tanks can be compared t o t h e emission l o s s pred ic ted by t h e AF'I

equa t ion , i t i s not f e a s i b l e t o make a d i r e c t comparison of t h e quant i -

t a t i v e e f f e c t of each v a r i a b l e o r f a c t o r o r t o a s s e s s t h e accuracy of

t h e va lues ass igned t o exponents or k f a c t o r s i n t h e API equat ion.

I n gene ra l , however, the ES l abora to ry experiments on g a s o l i n e

evaporat ion r a t e s , and t h e C B I j S O H I O p i l o t tank and o t h e r tank t e s t s

show t r ends of i nc reas ing emission r a t e s a s wind speed and s e a l gap

a r e a a r e increased .

I n t h e case of t h e ES f i e l d t e s t s t u d y d a t a , m u l t i p l e l i n e a r

r eg res s ion ana lyses were conducted using s tandard computerized s t a t i s -

t i c a l programs (CORRE/STPRG) t o examine t h e r e l a t i o n s h i p of t h e

measured o r observed va lues of t h e v a r i a b l e s l i s t e d i n Table V I - 1 t o

t h e measured emission r a t e s . I n b r i e f , the s t a t i s t i c a l a n a l y s i s program

conducts a mul t ip l e s tepwise l i n e a r r eg res s ion (STPRG) i n a fash ion

t h a t o r d e r s t h e v a r i a b l e s from greatest t o l e a s t c o r r e l a t i o n . The

program was exerc ised i n t h e logar i thmic mode t o produce a power

f a c t o r r e l a t i o n s h i p of t h e type:

Y = a x bl x b2 b3 --__ - 1 2 x3

where Y i s t h e dependent v a r i a b l e and x l , x2, e t c . , a r e indepen-

dent v a r i a b l e s . This form was s e l e c t e d s i n c e i n t h e MI-2517 equat ion ,

t h e emission r a t e 0.7 (Ly) w a s a func t ion of P O a 7 , and Vw . The r e s u l t s of t h e r e g r e s s i o n ana lyses were t h a t f o r t h e measured

f i e l d d a t a from t h e 13 g a s o l i n e s t o r a g e tanks , t h e emission rates

were more c o r r e l a t e d wi th t h e va lues of wind speed, RVP, ASTM d i s t i l l a t i o n f a c t o r , gap a r e a (Ag), and s tock temperature than with

number of gaps, t ank d iameter , and t h e o the r va r i ab le s . I n order of

h ighe r t o lower c o r r e l a t i o n f o r t h e f i r s t f i v e va r i ab le s :

1, = f (Vw, RVP, gap a r e a , ASTM d i s t i l l a t i o n f a c t o r , s tock femperaturc) Y

VI-?

ENGINEERING-SCIENCE - It is considered t h a t i t might be a s a p p r o p r i a t e t o compute TVP

or P f o r u s e i n the c o r r e l a t i o n a n a l y s i s i n s t ead of RVP and t h e ASTM

d i s t i l l a t i o n f a c t o r . In t h a t ca se , t h e c o r r e l a t i o n order of t h e most

s i g n i f i c a n t sets of t e s t v a r i a b l e va lues is:

The mul t ip l e c o r r e l a t i o n c o e f f i c i e n t f o r t h i s s e t i s approximately

0.8. However, an accu ra t e mathematical r e l a t i o n s h i p f o r t hese v a r i a b l e s

i n t h e form of :

is not obtained s i n c e t h e s tandard e r r o r of the r eg res s ion c o e f f i c i e n t

( t h e b exponents in t h i s case) f o r each v a r i a b l e i s s i g n i f i c a n t (due

t o t h e d a t a s c a t t e r ) and due t o power f a c t o r format of t h e r e l a t i o n s h i p ,

cons iderable unce r t a in ty results. This unce r t a in ty i s a r e s u l t not

only of t h e d a t a s c a t t e r but of t h e r e l a t i v e l y smal l number of test

tanks (13) compared t o the number of v a r i a b l e s (15).

I n a d d i t i o n , t h e r e s u l t i n g express ion cannot be compared d i r e c t l y

t o t h e API equat ion s i n c e A

i n an experimental r e l a t i o n s h i p t o L whereas i n t h e API equat ion ,

t h e gap f a c t o r i s i m p l i c i t l y assumed t o be a f ixed l i n e a r va lue f o r

each t a n k l r o o f l s e a l combination.

h e r e i s a continuous v a r i a b l e expressed g

Y'

T h e B a t t e l l e r e p o r t f o r t h e EPA ( see Bibliography) concluded t h a t

t h e API-2517 s t and ing s to rage loss equat ion was of ques t ionable accuracy.

More s p e c i f i c a l l y , a s t a t i s t i c a l eva lua t ion of t h e API wind speed and

vapor p re s su re d a t a s c a t t e r i n d i c a t e d t h a t the exponents of 0.7 f o r

both t h e V

reasonable c e r t a i n t y l i m i t s .

and vapor p re s su re f a c t o r s could not be confirmed wi th in W

The API emission l o s s equat ion was developed from d a t a compiled

on s to rage tanks 20 t o 40 yea r s ago.

l o s s e s from s t o r a g e tanks e x i s t i n g today is i n ques t ion i n view of

t h e previous d i scuss ion .

I ts a p p l i c a b i l i t y f o r es t imat ing

The d i r e c t comparison of measured emission

V I - 4

ENGINEERING-SCIENCE - r a t e s t o those p red ic t ed by API-2517 was shown in Figure V-1.

Ignoring t h e case of t h e Gulf O i l tank, t h e API equat ion p red ic t ed

l o s s e s ranging from one-third t o n ine t imes the l o s s e s measured by ES.

In seven cases , t h e API p r e d i c t i o n was s i g n i f i c a n t l y g r e a t e r , and i n

four cases , i t was s i g n i f i c a n t l y less. From t h i s r e s u l t , and without

a d d i t i o n a l d a t a , i t is considered t h a t t h e API express ion is not i n

t h e form t h a t w i l l produce reasonably r e p r e s e n t a t i v e e s t ima tes of

emission lo s ses . Add i t iona l ly , t h e incons is tency of t h e API pred ic t ions

compared t o t h e ES measured emissions make i t d i f f i c u l t t o i d e n t i f y

and eva lua te p o t e n t i a l causes f o r t h e d i f f e rences . It remains unclear

whether an emission lo s s r e l a t i o n s h i p can be expressed i n t h e form of

t h e API equat ion , or, i n view of t h e order of r e l a t i v e c o r r e l a t i o n

between t h e key f a c t o r s as i n d i c a t e d by t h e ES c o r r e l a t i o n ana lyses ,

how t o develop wi th less u n c e r t a i n t y t h e c o e f f i c i e n t s or exponents

f o r t h e v a r i a b l e s which should be included in t h e emission l o s s esti-

mation expression. Therefore , no f u r t h e r a t t empt s were made t o

develop an emission lo s s equat iop by s t a t i s t i c a l c o r r e l a t i o n ana lyses

or a n a l y t i c a l techniques.

An express ion f o r t h e evapora t ion r a t e of v o l a t i l e l i q u i d s i n t o

a i r has been der ived in t h e form:

NA = evapora t ion r a t e per u n i t l i q u i d su r face a rea .

DAB = d i f f u s i v e l y cons tan t f o r l i q u i d A.

P,T,R = a i r p res su re , temperature , and gas cons tan t .

z = stagnant a i r space he igh t .

PB2 = a i r p a r t i a l p r e s s u r e a t top of con ta ine r .

PB1 = a i r p a r t i a l p r e s s u r e a t su r face of l i q u i d .

VI-5

ENGINEERING-SCIENCE - using n-pentane ve r sus a r e p r e s e n t a t i v e petroleum l i q u i d :

2 2 Dm = 0.075 cm / s e c = 6 . 2 9 f t I h r .

PAl = 6.8 p s i a

PB2 = 1 4 . 7 p i a

(Pnl = 7 . 9 p s i a )

l b - molelhr ) . 2 ( .

4 . 6 5 ~ 1 0 - ~ Then, NA = f t z

Assuming t h e molecular weight of gaso l ine i s approximately 55 ' lb /

lb-mole, t h e z f o r tube s e a l s i n a f loat ing-roof tank is about two

f e e t , and the l i q u i d su r face exposed i n t h e r i m space is one-half f e e t

w i d e , then f o r a 100-foot diameter tank:

L = 4*65x10-4x 55 x 100 B x 112 x 24 = 48 lb lday Y 2

This r e s u l t i s wi th in a reasonable range of t h e va lues of

measured emissions from t h e 1 3 t anks (20 t o 265 lb/day f o r tank diam-

e t e r s ranging from 55 t o 153 feet) . Therefore i t i s considered t h a t

f u r t h e r eva lua t ion of t h e a p p l i c a t i o n of t h e t h e o r e t i c a l evaporat ion

equat ion t o t h e r e a l t ank s i t u a t i o n may prove t o be of va lue .

VI-6

CHAPTER V I I

BEST AVAILABLE SEAL TECHNOLOGY

.. .

ENGINEERING-SCIENCE - CHAPTER V I 1

BEST AVAILABLE SEAL TECHNOLOGY

GENERAL

P r i o r t o World War 11, most crude o i l and petroleum products with

vapor p re s su res l e s s than 11 o r 1 2 p s i a were s to red i n fixed-roof tanks.

However, t h e r e was some f loa t ing- roof technology a s e a r l y a s t h e 1920's.

It had been developed p r imar i ly t o conserve product.

are gene ra l ly up t o a f o o t or more smal le r i n diameter than t h e i n s i d e

of t h e tank t o a l low f o r v a r i a t i o n s i n t h e diameter of tank caused by

s e t t l i n g , wind load, d i f f e r e n t i a l expansion, r i v e t heads, b u t t s t r a p s ,

l a p welds, cor ros ion , e t c . Bulges i n tank s h e l l s a r e p a r t i c u l a r l y

preva len t i n tanks 100 f e e t or g r e a t e r i n diameter.

F loa t ing r o o f s

Two gene ra l techniques are used t o s e a l the annular space between t h e

f l o a t i n g roof and t h e tank s h e l l .

roof tanks had metal shoe s e a l s which pressed aga ins t the tank s h e l l and a

f l e x i b l e f a b r i c covered t h e space between the roof r i m and t h e top of

t h e shoe s e a l . More r e c e n t l y tube s e a l s ( f a b r i c covered foam o r l i q u i d

f i l l e d f a b r i c ) has been used t o s e a l the annular space.

When f i r s t introduced a l l f loa t ing -

The o b j e c t i v e of t h i s t a s k w a s t o determine by surveys t h e b e s t

a v a i l a b l e s e a l technology f o r f loa t ing- roof tanks , p a r t i c u l a r l y f o r

upgrading e x i s t i n g tankage where necessary. Major emphasis was placed

on seal technology (both primary and secondary) f o r pontoon-type and

double deck open f loa t ing- roof tanks . Both crude o i l and d i s t i l l a t e

tanks were s tud ied . Two types of surveys were conducted.-

The f i r s t survey of a v a i l a b l e tank seals cons i s t ed of personal

in te rv iews by te lephone and correspondence with both manufacturers

and users of f loa t ing- roof tanks and wi th tank e r e c t i o n con t r ac to r s ,

consu l t ing engineers , and governmental a i r p o l l u t i o n c o n t r o l engineers .

Various brochures and r e p o r t s of manufacturers and u s e r s y ie lded in-

formation of va lue a s d i d a t t endance a t s e v e r a l va r i ance hear ings

concerning seal technology.

t o i n s t a l l a t i o n s of i n t e r e s t . Names of o rgan iza t ions and persons

interviewed or providing information a r e l i s t e d i n Appendix E.

In a few cases , personal v i s i t s were made

V I I - 1

ENQINEERINQ-SCIENCE - In t h e second survey, i n s t a l l e d tank s e a l s were reviewed wi th

r e spec t t o gap s i z e . Over 200 t anks i n t h e South Coast A i r Basin

were analyzed. A da ta shee t f o r each tank , showing inspec t ion r e s u l t s

a t each roof l e v e l , appears i n Appendix F.

SURVEY OF AVAILABLE SEALS

Primary S e a l s

Primary s e a l s a r e designed t o prevent vapor l o s s from exposed

l i q u i d i n t h e annular space between t h e r i m of the f loa t ing- roof and

t h e tank s h e l l , u sua l ly about e i g h t inches. They may be mechanical

or tube type.

The mechanical s e a l s a r e v e r t i c a l shoes (metal p l a t e s about 30-36

inches i n v e r t i c a l dimension) connected by braces t o t h e f loa t ing- roof

around i ts e n t i r e circumference. The shoes a r e h e l d a g a i n s t t h e tank

w a l l by va r ious types of sp r ings o r weighted l e v e r s . The open space

between the top of t h e shoe and t h e rim of t h e roof i s c losed by a

s u i t a b l e coated f a b r i c . Thus t h e only por t ion of t h e l i q u i d i n t h e

tank which i s exposed t o t h e atmosphere is t h a t between t h e shoe and

t h e tank w a l l . In a p e r f e c t l y round, smooth s u r f a c e the gap between

the shoe and t h e tank s h e l l is almost n i l . In a c t u a l tanks imperfect ions

on t h e tank wa l l (e.g. . weld seams, r i v e t heads, b u t t s t r a p s , s o l i d

build-ups such as wax) or t h e l a r g e r imperfect a r e a s (bulges , d e n t s ,

out-of-roundness r e s u l t i n g from s e t t l i n g , wind, thermal s t r a i n , e t c . )

r e s u l t i n an i n c r e a s e i n gap s i z e f o r varying d i s t a n c e s around t h e

circumference of the tank. A l l major tank manufacturers supply

mechanical shoes of va r ious designs.

Figure VII-1.

A t y p i c a l des ign i s shown i n

The tube seal is a f l e x i b l e tube usua l ly f i l l e d with l i q u i d o r a

compressible s o l i d foam.

roof and t h e tank s h e l l u sua l ly a t o r above the l i q u i d l e v e l s o t h a t

they completely f i l l t h e annular space.

r e s i l i e n c e of t h e foam is such t h a t t h e s e a l can adapt i t s e l f t o wide

changes i n tank dimensions and even f i l l i n t o some ex ten t a t l e a s t ,

around pro t ruding r i v e t s and s h e l l imperfect ions.

The tubes a r e held between the r i m of t h e

The weight of t h e l i q u i d of the

V I I - 2

. F I G U R E V I I -

WEIGHT A S S I S T E D M E T A L L I C SHOE SEAL

ENGINEER ING-SCIENCE , I N C . . . VII-3

. ENQINEERINQ-SCIENCE m] - S e a l s o f f e red by t h e major tank manufacturers vary considerably

i n des ign d e t a i l .

( e .g . , nylon) coated with s y n t h e t i c e las tomer, r e s i s t a n t t o abras ion

and hydrocarbons. When a l i q u i d is used i n s i d e the tank tube i t is

commonly a petroleum d i s t i l l a t e or s i m i l a r m a t e r i a l t h a t w i l l not

contaminate tank con ten t s i n c a s e of a puncture.

The s e a l i t s e l f is usua l ly a re inforced f a b r i c

The foam s e a l most o f t e n is a n open-celled polyurethane foam.

Nearly a l l t ank manufacturers o f f e r one or more types of foam s e a l s .

They vary i n d e n s i t y , compress ib i l i t y , shape (round, hexagonal,

octagonal , e t c . ) and diameter (up t o 1 4 inches) . Bas i ca l ly , t h e foam

seal may be considered as a doughnut o r hoop-like log he ld between t h e

rim of t h e f loa t ing- roof and the tank s h e l l , e i ther p a r t i a l l y immersed

i n t h e l i q u i d i n t h e tank or in t h e vapor space j u s t above t h e l i q u i d .

Typical des igns are shown i n F igu res V I I - 2 and VII-3.

Major tank manufacturers o f f e r i n g primary tube s e a l s in Southern

Chicago Bridge and I r o n Company, American Bridge C a l i f o r n i a include:

Div is ion of U.S. S t e e l , P i t t s b u r g h - Des Moines Steel Company, and

GATX Tank Erec tor Corporation.

Severa l o the r manufacturers o f f e r or have developed s p e c i a l des igns

Chiyoda Chemical Engineering of primary seals similar t o t ube seals.

and Construct ion Company, Inc . o f f e r s a vapor s e a l i n g device b u i l t

i n t h e shape of a p a i r of l i p s and c a l l e d "Kisseal" (Figure VII-4).

The device is made of polyurethane foam bonded t o a cover shee t of

rubber-reinforced nylon wi th an a c r y l i c adhesive. The shape of t h e '

seal a l lows it t o a c t as a double seal wi th r e s p e c t t o small w a l l

deformations. The manufacturer claims t h a t whereas foam logs may

p a r t i a l l y f a i l by s p l i t t i n g when t h e annular space narrows, t h e design

of t h e "Kisseal" m i t i g a t e s such a c t i o n . The "Kisseal" has been in use

in welded tanks i n t h e Or ien t f o r t e n yea r s bu t has not been used on

r i v e t e d tanks. No i n s t a l l a t i o n s in t h e United S t a t e s are known.

During an in te rv iew wi th r e p r e s e n t a t i v e s of Chiyoda, it w a s learned

t h a t t h e ch ie f a t t r i b u t e of the s e a l , long l i f e , is due t o design of

t h e s e a l which m i t i g a t e s compressive f o r c e s , and thereby promotes

VTT-L

FIGURE VII-2

SR-7 HORTON FABRIC SEAL ( R E S I L I E N T TYPE)

FOR HORTON FLOATING ROOFS

WEATHER SHIELD-

HANGER BAR-

CURTAIN SEAL-

SEAL ENVELOPE -

SEAL SUPPORT- RING

RESILIENT URETHANE FOAM-

ENGINEER ING-SC IENCE, INC. VII-5

FIGURE VII.

=E W c v) > v)

I 0 n

~~ ~ ~

ENGINEERING-SCIENCE, INC VI1-h

/

l! LQ3 . . . . . .

E N G I N E E R I N G S C I E N C E , INC. V I I - 7

ENQINEER~NQ-SCIENCE [ESI - I r e t e n t i o n of e l a s t i c i t y . However, i n spec t ion of t e n tanks i n t h e

Orient by company personnel i nd ica t ed t h a t t h e s e a l d i d not meet t h e

C a l i f o r n i a A i r Resources Board c r i t e r i o n of no gaps g r e a t e r than one-

e igh th inch on e i g h t of t h e 10 t anks examined. I n s t a l l a t i o n co$t of

t h e seal i n tanks i n t h e United S t a t e s is expected t o be two t o t h r e e

times t h a t of domestic tube s e a l s .

The ThetaChron Divis ion of Joseph Weissenbach and Associates has

descr ibed a system f o r modifying a f loa t ing- roof tank t o e l imina te

vapor l o s s e s . The tank is f i t t e d wi th a f a b r i c top designed t o with-

s tand wind f o r c e s t h a t might o therwise blow i t o f f .

is f i t t e d wi th an e las tomer ic diaphragm s k i r t which e s s e n t i a l l y r ep laces

t h e tank w a l l (Figure VII-5).

t h e impervious s k i r t and t h e tank s h e l l is f i l l e d wi th e thylene g lyco l .

No commercial t ank of t h i s des ign has y e t been b u i l t .

Secondary Sea l s

The f loat ing-roof

The space between t h e o u t s i d e of

I n order t o prevent contamination of product by r a i n and dus t

smaller secondary s e a l s have been developed.

Wiper-type. The s imples t t y p e of secondary seal is t h e c l o t h

wiper a t t ached t o t h e top of t h e primary seal o r t h e top of t h e

f loat ing-roof r i m and extending t o t h e tank w a l l t o wipe of f

adhering l i q u i d . A diagram of a wiper w i t h three p ieces of f a b r i c

is shown i n F igure "11-6.

Other types of wiper seals being t e s t e d i n Southern C a l i f o r n i a

are:

U l t r a f l o t e . The u l t r a f l o t e wipe (F igure VII-7) has been used

f o r some time along wi th i n t e r n a l f l oa t ing - roo f s (used in covered

tanks t o reduce b rea th ing l o s s e s ) .

formulated polyvinyl c h l o r i d e shee t .

b racke t extending from t h e primary shoe is b u i l t up from t h e

i n s i d e wi th a foam material.

on gunni ted su r faces . I t is n o t recomended on r i v e t e d tanks.

It is made of a s p e c i a l l y

The end bo l t ed t o t h e

The wiper is non-abrasive and works

VII-8

FIGURE VII-

I t I 1

I

I

\ I

ENGINEER INC-SCIENCE. IN( VII-9

SECONDARY WIPER SEAL

a n W

9

W I- O -1 LL U z I- -1 3

- A Y

= w n

L - Y 8 0 Y

FIGURE VII-

ENGINEERING-SCIENCE, IN[ VII-11

L z I- A

Y 0

v) -

-I Y

ENQINEERINQ-SCIENCE - Mini-Bag. There a r e s e v e r a l t y p e s of mini-bag s e a l s . All a r e

small ve r s ions of primary foam tube s e a l s . They may be at tached

t o t h e top of the mechanical shoe or t o the roo f . Foam logs a r e

3 t o 4 inches i n diameter . Examples a r e shown i n F i g u r e s V I I - 8 ,

9 , and 10.

T r i co Wiper. T h i s i s a f l e x i b l e wedge wiper a t tached t o a bracket

bo l ted on the primary shoe and impinging almost perpendicular ly

on t h e wal l (F igure VII-11).

Maloney Sea l . The Maloney seal i s a f l e x i b l e w i p e r h e l d aga ins t

t h e tank wal l by means of a spr ing . Therefore , i t e x e r t s con-

s i d e r a b l e pressure . A diagram of t h e Maloney s e a l i s shown i n

Figure VII-12. The seal w i l l maintain contac t with t h e tank

wa l l , even when t h e shoe is 2 1 / 2 inches from t h e wal l .

Sea l and Tank Accessories

When a welded tank is new and in good condi t ion , p roper ly i n s t a l l e d

shoe s e a l s show a good f i t and few i f any gaps. In t i m e , however, the

mechanism fo rc ing t h e seal a g a i n s t t h e tank w a l l may d e t e r i o r a t e due

t o cor ros ion and build-up of s o l i d s allowing t h e shoe t o back away

from t h e tank wal l . To c o r r e c t t h i s problem, some u s e r s i n s t a l l

e x t e r n a l sp r ings on t h e r i m of t h e roof t o h e l p t h e top of t h e shoe t o

s t a y i n con tac t wi th the tank wa l l (F igu re VII-13). Too much p r e s s u r e ,

of course , may des t roy any coa t ing on t h e i n s i d e wa l l .

No p r a c t i c a l amount of p r e s s u r e on a m e t a l l i c shoe w i l l c l o s e gaps

caused by r i v e t heads i n r i v e t e d tanks. Accordingly, a t t empt s have

been made t o smooth t h e tank w a l l around r i v e t heads and b u t t straps

by f i l l i n g in t h e spaces between t h e heads and e l imina t ing abrupt

changes in t h e su r face of t h e t ank wa l l . Among f i l l e r s t h a t have been

used are gunni te , epoxy r e s i n s , and g l a s s f i b e r r e s i n mixtures . A l l

have t h e i r l i m i t a t i o n s and problems.

Weather guards and weather s h i e l d s a r e o f t e n i n s t a l l e d over

primary shoe and tube s e a l s . T h e s e t h i n me ta l p l a t e s which are pivoted

from t h e roof p r o t e c t any l i q u i d t h a t may be exposed by l a r g e gaps from

UV l i g h t and wind c u r r e n t s but do l i t t l e t o l i m i t gap s i z e . A t times,

VII-12

F I G U R E V I I -

M 0

A = Y > 0

u

m - - a L -4 - Y M

I- o W

W a

I I I I I I I I I I

I ;=

\ \

I I A I

V

z 0

I- U w w -1 w

-

z 0 - I- 4 > Y -1 Y

I- z 0 = L

ENGINEERING-SC I E N C E , IN1 VII-13

FIGURE V I I .

-3 4 rn

w M

;c 0

I- - > W 2 W

-

ENGINEER ING-SC IENCE, IN1 V I I - 1 P

FIGURE VII-l

u 0

-

-

4 I

t

r

z 0

c U > W 2 W

W

-

a - v)

z 0

c U > W _1 Y

c z 0 oz U

-

ENGINEERING-SCIENCE, INC. VII-15

FIGURE VII-1

I 2 2 Y I M

L z - c

ENGINEER ING-SC IENCE, INC. V I I - 1 6

MALONEY SECONDARY TANK SEAL

XISTING CLAMP

EXISTING PRIMARY SEAL

ENGl NEER ING-SC I ENCE, INC. VII-17

FIGURE V I I - 1

S P R I N G A S S I S T E D MECHANISM

ENGINEERING-SCIENCE, IN[ V I I - 1 8 . .

i

ENQINEERINQ-SCIENCE [as I - they have been considered by some as secondary s e a l s , bu t t h e i r e f fec-

t i v e n e s s in reducing gap s i z e is poor.

SURVEY OF INSTALLED SEALS WITH RESPECT TO GAP S I Z E

T e s t d a t a on more than 200 t anks were obtained from v a r i o u s oil

companies i n t h e South Coast A i r Basin (At lan t ic -Richf ie ld , Douglas,

Gul f , Mobil, S h e l l , Standard, Texaco, and Union). Se l ec t ion of tanks

and s e a l s t o be t e s t e d , methods of i n spec t ion , and of r epor t ing r e s u l t s

were made by each company i n d i v i d u a l l y .

from only one roof l e v e l up t o more than 20 d i f f e r e n t l e v e l s per tank.

Some inspec to r s repor ted gap measurements in ranges of 118" t o less

than 1/4", 114" t o 3/8", 318" t o 112". 112" and above, and m a x i m u m

gap s i z e . A l l companies repor ted d a t a in such form t h a t i t could be

arranged t o show gap measurements in two ca t egor i e s :

bu t less than 1/2", and (2) 112" o r more. A summary of t h e d a t a is

shown in Table V I I - 1 ( f o r welded tanks) and Table VI I -2 ( f o r r i v e t e d

t anks ) . Only t h e average of t o t a l gap l eng ths (average f o r t h e

numbers of roof l e v e l s where i n s p e c t i o n s were made) is repor ted in t h e

Tables in order t o avoid over-complicating them. Observations con-

cerning s p e c i f i c i n spec t ions a r e also shown. As te r i sks bes ide tank

numbers in t h e Tables r e f e r t o t h o s e in spec t ions where observa t ions

are p a r t i c u l a r l y p e r t i n e n t .

Inspec t ion frequency va r i ed

(1) 118" o r more

The ind iv idua l s h e e t s in Appendix F were der ived from t h e inspec-

t i o n r e p o r t s submitted by t h e v a r i o u s companies, and put i n t o a uniform format for c l a r i t y .

level, the ind iv idua l shee t w a s n o t included, bu t t h e r e s u l t s a r e

shown in Tables V I I - 1 and VII-2. Some of t h e r e p o r t s inc lude inspec-

t i o n s made by ARB and APCD personnel ; o t h e r s do no t . The in spec t ions

made by t h e agencies are not i d e n t i f i e d in t h i s r e p o r t .

Where inspec t ions were made only a t one

Tank i d e n t i f i c a t i o n numbers are meaningful only t o t h e au tho r s of

t h i s r e p o r t .

t h e data was preserved.

They were so arranged and s e l e c t e d so t h a t anonymity of

Examination of Table V I I - 1 ( f o r welded tanks) r evea l s t h a t only

one tank wi th a shoe seal and no secondary s e a l would meet a c r i t e r i o n

VII-19

ENQINEERINQ-SCIENCE - TABLZ V I I - 1

PERFORMANCE OF SEALS ON FLOATING-ROOF TANXS

IAv. per Lev.) Tank Diam. Primary Secondary Levels - l /a : ' - 112" & Iden. Serv ice ( f e e t ; S e a l Seal Tested i/2" Over

B-1

B-2

R-l* R- 2 R- 3

R-4 R-5"

R-6 R-7 R- 8 R-9 R-10 R-11 R-12 R-13 R-14 R-15 R-16 R-17 R-18* R-19

R-20 L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 M - 1

M- 2 M- 3

C.O. 160

C.O. 140

Gas 60 Av. Gas 60 Av. Gas 60

Av. Gas 60 Av. Gas 60

c -c 100

Reformate 150 Reformate 100 Gas 138 C.O. 100 Gas 119 C.O. 100 EFU 150 Pentane 67 Hexane 90 C.O. 155 Gas 110 Gas 110 Gas 110

C.O. 220 C.O. 221 C.O. 227 C.O. 242 C.O. 230 C.O. 230 C.O. 230 C.O. 260 C.O. 260 C.O. 200 Gas 54.5

5 6

Gas 54.5 Gas 54.5

Shoe

Foam Log Tube Tube Shoe

Tube Shoe

Shoe

Tube Shoe Tube Tube Tube Shoe Tube Shoe Shoe Shoe Tube Shoe Shoe

Shoe Tube Tube Tube Shoe Shoe Shoe Tube Tube Tube Tube

Shoe Tube

Looped Fabr i c None

None None Mini- Tube None Mini-

Maloney

None None None None None None None Elaloney Maloney None None Maloney Mini- Bag None None None None None None None None None None Weather- Guard None Weather-

Bag

19

16

9 9 7

7 1 2

10

13 4

13 3 8

12 1 4 4 3 6 7 5 4

15 2 1 24 3

15 1 4 17

4 4 7 4

4 3

3.1

24.5

4 . 1 3 .2 5.7

1.0 0.1

0

3.5 41.3 0.2 4.7 0.6

62.0 0 2 0

62.3 0.5

17.2 1 .3

72.6 64.0 76.7 0

93.7 115.3 228.9

0.03 4

15.6 13.1

0.4 1

0

24,. 4

4 . 1

0

0 0

0

0.4 0.25 0 0 0.03 8.0 0 0 0 9.3 0.01 0.6 1.0

13.3 18.7 23.5 0 13.3 14.9

9.2 0 0 2.0 0

0.5 3.3

.56

Guard

VII-20

ENGINEERING-SCIENCE [E s 1 -

TABLE V I I - 1

PERFCREUCZ OF SEALS ON FLOATING-ROOF TANKS !e) 'rleided Tanks

Feet of Gap (&I. p e r Lev.:

Tank D i m . F r i m a r y ' Secondary Leveis 118"- 112" & Iden t . Serv ice (fect) Sea1 Seal Tested 3.12'' Sver

M- 4 Gas

M-5 Diesel M- 6 C.O. M- 7 C . O . M-8 C.O. H- 1 Gas H-2 Gas H-3 Recov.

O i l H-4 Gas H-5* Gas H-5 Gas H-6 Gas H-?(a) Gas H-7(b) Gas H-8 Gas H-8(b) Gas H- 9 Ga 8

H-10 Gas H - 1 1 Gas H-12 Gas H-l3(R)* Gas H-13(W)* Gas P-1 C.O. P-2 Gas P-3 Gas P-4 ' Jet A P-5 Gas P-7 Av. Gas P-8 Rec. Gas P-9 Gas P-10 Gas P-11 P-12 P-13 P-14" P-15 P-16 P-17 P-18

54.5

21.5 60 90 90

120 150

80

120 120 120 117 1 1 7 117 117 1 1 7 120 120 120 120 117 1 1 7 1 1 7 120 122 122

96 60 48

120 30

Sour Water 75 C.O. 195 C.O. 160 Rec. O i l 45 Rec. O i l 25 Gas Oil 80 Gas 100 Alkylate 36

Tube

Shoe Shoe Shoe Shoe Tube Tube Shoe

Tube Shoe Tube Tube Tube Shoe Shoe Tube Shoe Tube Shoe Shoe Tube Tube Shoe Shoe Tube Tube Shoe Shoe Shoe Shoe Tube Shoe Tube Tube Tube Tube Tube Shoe Tube

Weather- 4 Guard None 3 Wiper 2 Wiper 2 Wiper 2 None 8 None 6 None 1

None 7 None 2 None 2 None 7 None 3 None 3 None 2 None 6 None 2 None 5 None 9 None 2 None 4 None 4 None 3 None 3 None 3 None 3 None 3 None 3 None 3 None 2 None 4 None 3 None 5 None ' 4 None 3 None 3 None 3 None 3 None 3

5.8

7.5 34

162 142

0.34 0 7 . 2

2 . 2

0 10.1 0 3.3

11.4 0 3.8 0 8.3

15.3 13.2 0

10.8 14.0 0.9 2.4

18.4 11.8 14.1 36.2

10.1 0.4 0.3

89 5 0.5

42.0 4.7

.13

.08

0.4

0 2.9

7 1 55.9 0.3 0 0

0 0 .04 0.08 3.4 0 0.9 5.4 0 0 0.07 0 0

11.1 0

31 0 0.01 0.001 7.7 0.47 0.7 0.1 0.2 0.9 0 0 3.1 0.4+ Dt 7.5 0.8

V I I - 2 1

ENGINEERING-SCIENCE IES] -

TABLE V I I - 1

PERFC.W?CE OF SEALS ON FLOATING-ROOF TAhXS (Cont.)

Welded Tanks Feet of Gap

(Av. per Lev.) Tank Xam. Primary Secondfry Levels :/a”-. 1 / 2 “ 6 Iden t . Serv ice ( f e e t ) S e a l Sea l Tested 112‘‘ Over

P-19 c-1

c-2 c-3 c-4

c-5 C-6 c- 7 C-8 c-9 c-10 c-11 c-12 C-13 C-14 T-1 T-2 T-3 T-4

T-5

T-6

T- 7

T-8

T-9

T-10 T-11

Alkylate 100 Av. Gas 95

Gas 140 Gas 140 Gas 140

Gas 30 Gas 86 Gas 78 C.O. 7 8 Gas 94 J e t 120 Naphtha 110 Sour Water 140 Gas 140 C.O. 140 Naphtha 90 Naphtha 90 Naphtha 120 Gas 120

Gas 120

Naphtha 120

Gas 120

Alky le t e 120

Reformer 150 Feed Reformate 150 JP-4 150

T-12 Reformate 150 . T-l3(a) Gas 120 T-13(b) Gas 120 T-14 Resid. 120

Stk. T-15 Recov. o i l 90 T-16* Gas 120

Shoe Shoe

Shoe Shoe Shoe

Tube Tube Tube Shoe Shoe Tube Shoe Shoe Shoe Shoe Shoe Tube Shoe Tube

Tube

Shoe

Tube

Tube

Shoe

Shoe Shoe Tube Tube Tube Shoe

Shoe Shoe

None Ultra-Type

Blade (Ex) Foam Sea l Ultra-Type (Ex) None None None None None None None Wiper Wiper None None None None Weather Shie ld Weather Shie ld Weather Shie ld Weather Shie ld Weather Shie ld None

None None None None Maloney None

None None

(EX)

4 3

1 4 3

1 2 3 1 1 1 1 1 1 1 1 1 1 1

2

2

3

1

1

1 1 1 2

All 1

1 3

11 0

5.6 0.9 3

0 0.05 2.7 3.3

12 .1 0

27.2 2 .2 1 .5 2.1

59.4 0

53.1 55.8

17.2

0

8 .8

4.6

94

102.9 56.9 0 7.0 0.1 5.7

0 3.4

0.2 0

6.7 0 0

0 3.4 0 . 0.2 0 0 0

33 0 0.6

0 8 .3

77.4

0.7

0

0.6

0

1 2

46.8

4.6 0 .5 0 6.2 0 0

0 0.07

VI I -22

ENQINEERINQ-SCIENCE IES] -

TABLE V I I - 1

PERFOREMCE OF SEALS CN FLOATING-WOF TANKS (Cont .)

Welded Tanks Feet of Gap

(Av. per Lev.) Tank C i a m . Primary Secondary Levels l / P - i:2" & Iden. Se rv ice ( f e e t ) Seal S e a l Tes:ed 1/2" Cver - T-17 T-18

T-19 T-20 T-21 T-22 T-23 T-24 T-25 T-26

T-27

Y-l*

y-2*

Y-3

Y-4"

Y-5

Gas 120 Gas 120

Recov. O i l 52 Av. Gas 60 Av. Gas 60 Ga s 120 Recov. O i l 60 Gas 90 Gas 70 Gas 150

Reformate 120

JP-4

JP-4

JP-4

Gas O i l

C.O.

60

60

90

86

150

Tube Shoe

Shoe Tube Tube Tube Tube Tube Shoe Tube

Tube

Foam Log Foam Log Foam Log Foam Log Liquid- F i l l e d Fabr i c

None 1 Weather 1 Sh ie ld Wiper 1 None 3 None . 2 None 3 None 3 None 2 None 2 Weather 4 Shield Weather 1 Shield None 7

None 2

None 5

None 6

None 4

2.2 1 . 4

1 . 7 7.1 .9 1.2 1 . 4 5.3

3.5

0

4.1

0

0

12.3

0

28

0 0

0.2 2.5 0 0 0

17.2- 1.3 0

0

30.4

0

2.9

56.8

8

VII-23

CNOlNtERINO-SCIENCE PI-

TABLE UII-2

P E R F O W W E OF SELLS ON FLOATING-ROOF TANKS

Riveted Tanks Feet of Gap

(Av. p e r Lev.) Tank Diam. Primary Secondary Levels 113"- 112" & Iden. Serv ice (Feet) Seal Sea l Tested 112" Cver

A- 1 K-1 K- 2 K- 3

K-4 K- 5 K-6* K-7* K-8 K- 9 K-10 K - 1 1 F-1 F-2 F-3 F-4 F-5 F-6 F-7 F-8 F-9 F-9.1 F-10 F-11 F-12 F-13 F-14 F-15 F-16 F-17 F-18 F-19 F-20 F-21 s-l* s-2 s-3 s-4 s-5 S-6

C.O. 135 Benzene 36 Naphtha 60 Recov. 115 Oil Gasoline 115 Benz-To1 115 Benz-To1 115 Benz-To1 115 Alkyla te 120 Gasol ine 115 Reformate 115 C.0 . 115 Gasoline 144 Gasol ine 144 Gasol ine 120 Gasol ine 120 Gasol ine 120 Gasol ine 120 Varied 120 Gasol ine 120 Gasol ine 120 Gasol ine 120 C.O. 144 C.O. 144 C.O. 144 C.O. 144 C.O. 144 C.O. 14C. C.O. 168

C . O . , 168 C.O. 168 C.O. 168 C.O. 168 Gasol ine 50 Rec. O i l 36 C.O. 117 C.O. 117 Gasol ine 117 Gasol ine 117

C.O. 168

Shoe Tube Shoe Shoe

Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe

Viper None Mini-Bag None

Mini-Bag Mini-Bag Mini-Bag Mini-Bag Mini-Bag None Mini-Bag None None None Wiper Wiper Wiper Wiper Wiper Wiper Wiper Wiper None None None None None None None None None None None None Foam None None None None None

18 3

16 4

5 3 4 2 8 2 5 3 4 4 4 4 4 4 3 4 4 4 3 3 3 3 3 3 3 3 3 2 2 2 4 3 3 3 4 2

23.55 0.75 0.24

47.5

5 . 1 4.2

29.3 19.0 0

15.5 12.3

146.7 86.7 31.2

130 148.9 105.8 89.8 60.8 75.6 70.3 45.1

201.8 138.7 200.'2

73.2 188.4 106.7 115.8 91.9

136.4 90.7

115.1 173.8

4 . 7 11.7

158 .O 67 . 3 85 .o 6 1 . j

8.1 0 0 7.5

0 2.7 3.5 3.3 0 - 3.0 C 4.7

147.9 30.2 45.4 27.1 52.9 14 17 .1 32.4 45.6 65.8 58.9 38.7 40 .O

198.3 43.6 56 . 4

242.2 56.7 50.7 69 .O 67.6 94.8 3.9

14.3 0

Unknown 3.9 0

VII-24

c

LNQINEERINQ-SCIENCE (E61 -

TABLE V I I - 2

PERFORMANCE OF SEALS ON FLOATING-ROOF TANKS (Cont.)

Riveted Tanks Feet of Gap

(Av. per Lev.) Tank Diam. Primary Secondary Levels 118"- 112" & Iden. Serv ice (Feet) S e a l S e a l Tested 112'' Over

5-7 S-8 s-9 s-10 s-11 5-12* S-13 5-14" S-15 S-16 S-17 s-18 x-1 X- 2

x-3 x-4 x-5 X-6 x-7 x- a x-9 x-10 x-11 x-12 X-13 X-14 X-15 X-16 X-17

x-19 .v-1 V- 2 v-3 V-4 v- 5 V- 6 V-7X

v- s

x-ia

Gasol ine 117 Rec. O i l 1 1 7 Gasol ine 1 1 7 Gasol ine 1 1 7 Gasol ine 1 1 7 Av. Gas 79 Av. Cas 60 Av. Gas 35 Gasol ine 1 1 7 Gasol ine 1 1 7 Gasoline 117 Rec. Oil 45 Gasol ine 1 4 4 Gasol ine 144

Jet 60 Av. G a s 60 Av. Gas 60 Av. Gas 60 Av. Gas 60 Jet ,117 Jet 117 Gasoline 120 Gasol ine , 1 2 0 Gasol ine 144 Gasol ine 144 C.O. 144 C.O. 144 C.O. 144 C.C. 144 C.O. 144 C.O. 1 4 4 Recov. O i l 96 C.O. 117 C.C. 117 C.C. 1 1 7 C.O. 1 1 7 Gasol ine 117 ?.esid. 120

C.O. 144

Shoe Shoe Shoe Shoe Shoe Shoe Tube Shoe Shoe Shoe Shoe Shoe Shoe Shoe

Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Shoe Skoe S h e

Shoe

None None None None None Foam Tube None Foam Tube None None Loop '

None F,inger (Ex.) Multi-Finger

None Wiper None Wiper Wiper Wiper Wiper Wiper Wiper Wiper None None None None None None None Sone None Hone X p e r None Uiper

Kini-Tube None

(EX.)

!Jith/W. 0.

3 2 2 2 2 5 4 4 3 4 4 3 1 2

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2

1

64.8 27 11.8 10.5 27 3 .1 -3.3 2

50 79 20 2 1 36.2

4

39.6 44.4 46 13.5

7 . 2 0 0

38.2 97.9 25.5 3

87.1 15.4 21

5.5 0 5.5

a6.5 159

2 54.2 33.5 3 1

123.7 1

28.9

1 . 2 0 0 1 .9 6.5 0.8

10 17.5

4.3 9 . 4.5 0.7 0 5.8

52 7 .2 0 0 0.3 2 0 0 0 0

248 0.2 0.2 0 0.2 0 4

14.8 12.2

4.2 1 . 7

20.3 0.6

59.6 0 0.8

ENQINEERINQ-SCIENCE -

TABLE V I I - 2

PERFORMANCE CF SEALS ON FLOATING-RO@F TANKS (Cont.)

Riveted Tanks Feet of Ga?

(Av. pe r Lev.) Tank D i a m . Frimary Secondary Levels 1:8"-' l i 2 " & Iden. Se rv ice ( f e e t ) S e a l S e a l Tested 1:2" Over

v-9* C.O. 144 Shoe WithlW.0. 2 214.6 2.5 Mini-Tube 63.4 51.7

v-10 C.O. 144 Shoe None 1 167.7 0 v-11 C.O. 144 Shoe None 1 140.4 3.9 v-12 V-13 V-14 v-15 V- 16 V-17 v-18 v-19 v-20 v-21 v-22

Alky la t e Gasoline Gasoline Gas Solv. Gas Solv. Gas Solv. Gasoline Addit ive Gasoline Gasoline Gas Solv.

120 27 2 3 27 27 27 27 30 30 30 30

Shoe Shoe Shoe Shoe Shoe Shoe Tube Shoe Shoe Shoe Shoe

Wiper None Wiper None None Wiper None Wiper None None Wiper

37.7 2 . 1 40.4 0.8

0.9 0 49.0 4.2 36.4 1.3 13.7 1..8 0 0 6.9 0 -

64.3 0.4 48.5 1 . 9

6.4 0 V-23 Gas Solv. 30 Shoe Wiper 1 1.9 0

VII-26

CNGINLERINQBCILNCL - of "no gap g r e a t e r than 1/8-in."

inspected only a t one roof l e v e l .

t o T-16. It met t h e c r i t e r i o n , bu t again, only a one-level i n spec t ion

was made. Several o t h e r tanks came c l o s e , b u t even t h e s e had gaps in

excess of 1/2-inch.

Th i s was Tank 15 and i t had been

Af ter an e x t e r n a l sp r ing was a t tached

When tanks w i t h primary shoe s e a l s were provided w i t h c e r t a i n sec-

ondary s e a l s , improvement was noted b u t the data a r e obviously l imi t ed .

Secondary s e a l s of fabr ic -pro tec ted foam brought t h e tanks wi th primary

shoe s e a l s i n t o a low-gap regime. The Maloney secondary s e a l and t h e

experimental u l t r a - type s e a l a l s o showed promise. Ordinary wiper s e a l s ,

however, d id l i t t l e good.

The primary tube s e a l performed much b e t t e r i n meeting t h e no-gap r u l e

Ehan t h e m e t a l l i c shoe seals. For tanks i n c l ean s e r v i c e provided with

primary tube seals, t h e number and s i z e of gaps found by in spec t ion were

much l e s s than wi th shoe s e a l s , a l though q u i t e a few were in excess of

1/2-inch. Of t h e primary tube seals t h a t f a i l e d badly, most were i n s t a l l e d

on crude o i l tanks. Weatherguards o r weathershields d id not improve t h e

performance of primary tube s e a l s .

Examination of Table V I I - 2 ( f o r r i ve t ed tanks) shows t h a t only

t h r e e of t h e tanks had primary foam-tube s e a l s . A l l of t h e r e s t had

primary m e t a l l i c shoe s e a l s . This r e f l e c t s t h e low l i f e expectancy

of tube seals due t o wear and t e a r of t h e f a b r i c envelope as r u s t and

o t h e r d e p o s i t s bu i ld up around r i v e t heads, b u t t s t r a p s , and overlapping

courses .

Only one r i v e t e d tank (K-8) showed no gaps i n excess of 1/8-inch a t

any level t e s t e d . However, t h i s tank is l i n e d wi th gunni te and no r i v e t s

are pro t ruding t o create s e a l gaps; thus i t e f f e c t i v e l y r ep resen t s a

s u r f a c e area similar t o t h a t of a welded tank. Other r i v e t e d tanks f i t t e d

wi th primary shoe s e a l s and mini-bag (smal l foam logs) secondary seals

showed a minimization of gap s i z e , b u t t h e gaps usua l ly included some i n

excess of 1/2-inch. As was t h e case wi th welded tanks , t h e u s e of sec-

ondary wipe s e a l s does not appear t o add anything t o gap minimization.

The t h r e e tanks with primary tube s e a l s only were a l l small.

Only one showed no gaps, and t h i s was measured a t one l e v e l only.

ENGINEERING-SCIENCE - DISCUSSION OF RESULTS

This d i scuss ion of f a c t o r s a f f e c t i n g s e a l performance is based

p a r t l y on t h e t e s t d a t a presented in Tables V I I - 1 and V I I - 2 , and

p a r t l y on t h e experience of f loa t ing- roof tank manufacturers , e r e c t o r s ,

and use r s . The d i scuss ion is d i r e c t e d s o l e l y t o t h e minimization of

seal gaps and c a r r i e s no connota t ion a s to hydrocarbon loss by evapor-

a t i o n .

The da ta l i s t e d i n Tables V I I - 1 and V I I - 2 showed t h a t i n p r a c t i c e

f loa t ing- roof tanks wi th m e t a l l i c shoes a s primary s e a l s d i sp l ay gaps

of va r ious l eng ths along t h e tank circumference and up t o seve ra l

inches between the s e a l and t h e s h e l l . The dimensions of these gaps

are r e l a t e d t o s e v e r a l real-world f a c t o r s t h a t should be understood

t o a s s e s s b e s t a v a i l a b l e s e a l technology.

The number and s i z e of t h e gaps depend on t h e symmetry and con-.

d i t i o n of t h e tank s h e l l , t h e s t a b i l i t y of t h e roo f , and t h e condi t ion

of t h e m e t a l l i c shoe and i t s opera t ing mechanism, a l l of which a f f e c t

t h e c learance between t h e seal and t h e s h e l l . I f each of t hese f a c t o r s

is e s s e n t i a l l y as designed, t h e r e should be no gap exceeding, perhaps,

1/8-inch.

D e s p i t e good design, t h e r e a r e a v a r i e t y of cond i t ions t h a t may

develop a s f i e l d condi t ions vary and time passes . For in s t ance , i f

t h e tank becomes out-of-round (e .g . , tending toward egg-shape) even

temporar i ly , or becomes bulged o r dented, the shoe must have t h e

c a p a b i l i t y of conforming t o t h e new contour i f l a r g e r gaps a r e t o be

avoided.

of t h e foundat ion o r t h e ground under t h e foundat ion, from thermal

stresses ( d i f f e r e n t i a l hea t ing by t h e sun), or from wind pressure .

Bulges may r e s u l t from thermal o r phys ica l stresses dur ing e r e c t i o n .

Tanks of l a r g e diameter a r e more prone t o these v a r i a t i o n s than smaller

tanks.

cu r ren t study.

Tank "out-of-roundness'' may r e s u l t from d i f f e r e n t i a l s e t t l i n g

Tanks up t o 260 f e e t in diameter were inves t iga t ed during t h e

V I I - 2 8

ENQINEERINQ-SCIENCE - If the roof shifts to one side of the tank or another, the shoe

seal must be capable of accommodating the large gap that forms on one

side of the tank, and must do this fairly uniformly. Roof shifting

often results from twisted roof poles or roof drains with stack swivel

joints that tend to pull the roof off-center, particularly at lower

levels.

shift, any imperfection in the tank wall such as rivet heads, butt-

straps, beaded welds, guide bars, or a build-up of corrosion products,

carbon, or heavy wax will hold the shoe sufficiently away from the

wall to produce gaps. Bulges or dents in addition to surface imper-

fections accentuate gap development.

Even if the tank is perfectly round and the roof doesn't

With all of the foregoing factors tending toward changing the

condition and even the dimensions of the tank over the years, it is

not surprising that gaps develop.

To mitigate some of these problems, the foam seal was developed. Theoretically, the foam seal will adjust itself to unevenness in the

surface of the tank shell, even to areas between rivet heads. The less dense the foam and the softer its plastic cover is, the more

readily it will fit itself to small imperfections in the wall surface.

On the other hand, the plastic envelope must be hard enough and thick enough to withstand abrasion and cutting; the foam must be sufficiently

dense to withstand compression and retain elasticity.

results and the foam seal loses some of its potential for perfect

sealing and some of its potential for long life.

seal fails it is either because it becomes torn and allows the foam

to become oil-soaked, or high compressive forces destroy its elasticity.

A tendency to form wrinkles as it is forced out-of-round leads to gap

formation.

So a compromise

When a foam tube

To cover up any gaps that occur in the primary seal, one may

install a secondary seal, hoping the gap in the two seals may not

line up vertically. A good secondary seal should be either softer or

harder than the primary seal with the hope that the different type of closure will add to the overall effectiveness of the dual seals. The

secondary seal should also have the potential to reach out further than

the primary seal and thus cover up large gaps.

VII-29

ENQINEERINQ-SCIENCE - The most common types of secondary seals are the foam or liquid

filledtube seals, the solid "high pressure" seal, and the clinging

wiper seal-. The secondary tube seal is usually considerably smaller

(less diameter) than the primary Lube seal, and may have a softer

envelope. The smaller diameter, the greater softness and the greater

throw" (potential to extend further than the primary seal) make it

effective when used in conjunction with either a shoe or tube seal as

the primary one. It is, of course, subject to the same tearing, and the

potential to wrinkle and form gaps as the primary tube seal.

!,

Solid secondary seals include the spring-loaded Maloney seal and

the solid elastomer seal of the ultra-type. Both of these seals

function by the extra pressure they exert on the tank wall, their

ability to fit into smaller tank bulges, and their greater "throw".

Although the pressure exerted is an advantage to the seal, it is also

a defect in that it may catch on protrusions such as rivet heads and

be ripped off, particularly when the roof is ascending; it may also

destroy any coating on the tank wall. appear to be adapted to riveted tanks.

Hence, the Maloney seal does not

The wiper seal has the advantage of long "throw" and some of the

softness properties of the tube seal. It probably is somewhat cheaper.

"Best available seal technology" implies inclusion of economic

factors such as original cost, maintenance costs, and seal life in its

determination. Insufficient information was available to make this

general decision.

The age, construction, and condition of floating-roof tanks varies greatly. It must be emphasized therefore, that the best method of minimizing gaps for one tank is not necessarily the best for another.

Riveted tanks display gaps of the same magnitude as the dimensions of

surface protrusions (e.g., rivet heads, butt straps, course overlaps)

when they are impinged upon by even the most effective primary seal.

Secondary foam tube or similar seals will minimize the number and

size of gaps that line up vertically in a riveted tank.

not riveted tanks may be modified (by gunnite or other coatings)

Whether o r

VII-30

ENQINEERING-SCIENCE - sufficiently to eliminate gaps has not been demonstrated.

methodology for minimizing seal gaps in riveted tanks appears to be a

primary shoe seal plus a secondary tube (or similar) seal. In order

to meet gap minimization regulations even with this "best" methodology,

exemptions for protrusions inherent in tank construction (rivet heads, etc.) are required.

The best

Perfectly round (or nearly s o ) welded tanks in clean service

may meet a "no gap greater than 1/8-inch" criterion with primary shoe or tube seals only. Many welded tanks with primary seals only, or a

combination of primary and secondary seals may show "no gaps greater

than 1/8-inch" at many roof levels and still show limited gaps up to

l/Z-inch at a few other levels. This is apparently due to stresses

that developed during erection, or by environmental changes during

use. The best method of minimizing seal gaps f o r welded tanks appears

to be either a primary shoe or tube seal plus a secondary seal of the

tube o r pressure type. Even so, gaps up to one-half inch may appear

at some roof levels depending on the configuration of the specific tank.

VII-31

CHAPTER V I I I

RECOMMENDAT IONS

? ?

-, . . ' . ..

, . . . . . i : . .

ENQINEERINQ-SCIENCE - CHAPTER V I 1 1

RECOMMENDATIONS

1.

2.

3 .

4 .

Seal gap s i z e should not be over-emphasized a s t h e c o n t r o l l i n g

v a r i a b l e ( t o the exclusion of o t h e r v a r i a b l e s ) t h a t may a f f e c t

hydrocarbon emission r a t e s from f loa t ing- roof s to rage tanks.

Further research should be conducted t o i d e n t i f y t h e r e l a t i v e

importance of t h e many p o t e n t i a l v a r i a b l e s . Such r e s u l t s would

poin t the d i r e c t i o n f o r developing new c o n t r o l technology.

Best a v a i l a b l e s e a l technology should not be def ined without

having s u f f i c i e n t emission d a t a t o v e r i f y t h e reduct ion in

emissions a t t r i b u t a b l e t o t h e u s e of a given s e a l on a s p e c i f i c

app l i ca t ion . Without emission measurements d a t a , conclusions

may be drawn t h a t would r e q u i r e emphasis on t h e wrong parameters

t o reduce hydrocarbon emissions.

The API Method f o r measuring hydrocarbon l o s s e s from f l o a t i n g -

roof s t o r a g e tanks (MI-2512) should be modified t o inc lude

sampling and a n a l y t i c a l procedures s i m i l a r t o those developed

during t h i s i n v e s t i g a t i o n .

The API Method f o r e s t i m a t i n g , by c a l c u l a t i o n , t h e hydrocarbon

l o s s e s from f loa t ing- roof s t o r a g e tanks (API-2517) should be

reviewed and updated t o r e f l e c t c u r r e n t technology i n tank, roo f ,

and s e a l designs. The updated b u l l e t i n should inc lude da ta on

double-deck f l o a t i n g r o o f s and v a r i o u s types of secondary s e a l s .

~

.

V I I I - 1

CHAPTER IX

B I EL I OGRAPHY

LNOINEERINQ-SCIENCE - CHAPTER IX

BIBLIOGRAPHY

"Compilation of Air Pollutant Emission Factors AP-42," U.S. Environmental Protection Agency, February 1976.

"Evaporation Loss from Floating-Roof Tanks," API Bulletin 2517, February 1962.

"Evaluation of Methods for Measuring and Controlling Hydrocarbon Emissions from Petroleum Storage Tanks," Report to U . S . Environmental Protection Agency by Battelle, Columbus Laboratories, Columbus, Ohio, November 1976.

"Floating Roof Emission Test Program," Report to Standard Oil Company (Ohio) by Chicago Bridge & Iron Company, November 1976.

Gray, Douglas C. "Solvent Evaporation Rates," American Industrial Hygiene Association Journal, November 1974.

Gilbert, Theodore E. "Rate of Evaporation of Liquids into Air." JOUKnd of Paint Technology, November 1971.

"Mobil Oil Corporation - Floating Roof Tank Evaporative Loss Study," Preliminary Report to California Air Resources Board Workshop, December 17, 1976.

"Revision of Evaporative Hydrocarbon Emission Factors." Report to U.S. Environmental Protection Agency by Radian Corporation, Austin, Texas, August 1976.

"Standard Method of Test for Density and Specific Gravity of Liquids by Bingham Pycnometer," ASTM Designation: D-1217-54, American Society for Testing of Materials.

Standard Oil Company of California, Western Operations, Inc. Report for Public Hearing, June 25, 1976 to California Air Resources Board.

"Storage of Volatile Liquids," Chicago Bridge and Iron Company Monograph.

"Tentative Methods of Measuring Evaporative Loss from Petroleum Tanks and Transportation Equipment," API Bulletin 2512, July 1957.

"WOGA/CBI Floating Roof Emission Test Program" - Interim Report to Western Oil and Gas Association by Chicago Bridge and Iron Company, January, 1977.

IX-1